Hour1

• Raghupathi NarayanaRao Cavale

• [email protected]
• cnraghupathi@manipal,edu
• +91-9980540738
Trading
How Algo trading came…history
• The evolution of trading is one of the most significant factors in
the journey of mankind. Humans have evolved throughout the
centuries and that would not have been possible if they were
restricted to geographical boundaries.
• Trading is a system of bringing people together for mutual
benefit, though the primitive societies saw gatherings of people
only in religious and cultural events which were limited by
custom or kinship.
• Let us look at the hisorty
Stone Age
The Stone Age began roughly 2.6 million years ago; during this
era the stone tools were used for hunting and people were self-
sufficient. They traveled in search of food and shelter, trade was
conducted in relatively smaller scale, within small communities,
and over a shorter distance.
There was no concept of farming and merchants in the early
Stone Age (Paleolithic stage of the Stone Age). Trading was the
main facility of prehistoric people, who bartered goods and
services including hunting equipment and stones considered to
have great value.
Currency Used: Barter system
Obsidian
17000-9000 BC
Obsidian was a widely used trade item post 17,000 BC because of the great desirability of
this material before the use of metals. Obsidian was used to make cutting tools and was
preferred over other available materials since it also signified the higher status of the tribe.
Obsidian was traded at distances of 900 kilometers within the Mediterranean region.
With the introduction of the new Stone Age era (Neolithic Phase) i.e starting 9000 BC
agriculture started developing, people started domesticating animals and growing crops
which led to communities settling in one location.
With agriculture and new farming tools, there was more surplus of food which was used for
trading in exchange for other useful commodities (Barter System). Trading started between
different communities where not only surplus of food but also farming tools (tools made from
stone) and crafts were exchanged. With this, a new social class of merchants came into
existence; they would travel thousands of miles on foot to conduct trade with other
communities.
Currency Used: Commodities in the form of livestock, salt, metal, rare stones etc
8000-6000 BC
• It’s 8000 BC and the world population is around 5,000,000. People have now learned the art of
farming, domesticating animals, agriculture is booming and a food-producing economy has
emerged.
• Pottery traditions were on the rise in parts of the world now known as Asia, Japan, Korea, China,
Mexico and many more. Obsidian was still an important part of trading habit but there was more to
trade (barter) now including livestock, surplus production, salt, copper, cowry shells, pottery, animal
skins, farming tools, seeds etc.
• In the next few centuries’ people started inhabiting other parts of the world including Indus Valley,
Jordan, Ireland, Anatolia, Scotland, North America, Nigeria, Turkey, Norway, Italy, Europe, Egypt
etc. Ornaments (Gold & Copper) came into existence and were in huge demand around the world.
• Based on the inhabitant region these cultural group of people started accommodating necessities
based on the availability of natural resources in the adjacent areas. Objects were now made with an
aesthetic value and not just limited to function.
• Newly settled people now started importing exotic goods over distances of many hundreds of miles.
• Currency Used: Money in the form of objects such as weapons, metal artifacts, pottery, copper,
plant produce.
5,000 BC to 4000
BC

• Important Inventions:
The fast & Chronicle
The invention and development of ‘Wheel’ is an important part of the
trading history. This civilization that witnessed the use of wheel had a
better advantage over their ancestors because they had a greater
ability to produce food, manufacture goods, and transport people and
goods at a greater distance.
Communities started expanding since it became easier for them to
cover greater distance within a short period of time and there was no
need to stay close to food production areas. The wheel was also used
in making pottery; it was an important part of early civilization.
The period from 5th Millennium to 4th Millennium is also important
because this is when proto-writing or the first form of writing was
discovered. People were now able to commute and communicate
better.
 Clay amulet, one of
the Tărtăria tablets,
dated to ca. 4500
BCE

• Currency Used: Barter, clay

pots or tools, seeds, grains,
tools etc
Caravans of India: 3000 BC to 1000 BC
Ebla (Syria) became a prominent trading center in the third millennium. Trade
flourished between communities that were part of different kingdoms. With more
inventions acting as a catalyst for trade related activities the flow of goods
increased and the new set of occupations were acquired.
By the second millennium BC, former backwater island Cyprus had become a
major Mediterranean player by ferrying its vast copper resources to the Near East
and Egypt, regions wealthy due to their own natural resources such as papyrus and
wool. Phoenicia, famous for its seafaring expertise, hawked it's valuable cedar
wood and linens dyes all over the Mediterranean. China prospered by trading jade,
spices and later, silk. Britain shared its abundance of tin.
With the domestication of camels, trade routes over land became popular and the
group of traders called caravans used these trade routes to carry trades with India
and Mediterranean. Towns began sprouting up like never before with pit-stop or
caravan-to-ship port emerging everywhere.
It was during this time that the Incense Route was used to transport frankincense
and myrrh which was used as oil, perfume, incense, and medicine, which was only
found in the southern end of the Arabian Peninsula (modern Yemen and Oman).
Currency Used: Gold, silver, bronze, cattle, cowry shells, salt, commodity
exchange etc
•
Trading Routes 1000 BC….. 1000AD
• The city of Vulci became the hub of trading and manufacturing center.
• Greek colonists set up trade centers at Salona. The establishment of colonies permitted
the import and export of luxury goods such as pottery, wine, oil, metalwork, and textiles.
• The Han Dynasty opened up the 'Silk Road' trading route between China and Central
Asia. Many different kinds of merchandise traveled along the Silk Road, it is one of the
oldest routes of international trade in the world.
• The first non-stop voyages from Egypt to India were initiated at the start of the Common
Era. Spices from India became famous around the world and were the main exports to the
western world. The spice trade fostered new diplomatic relationships between East and
West, it was partly the spice trade in mind that Christopher set out in 1492 and ended up
finding America)
• With the start of 7th century AD the ‘Tea Horse Road’ was used to trade Chinese tea and
Tibetan warhorses. This route was spread over 6000 miles and was majorly used to
export tea from China to Tibet and India.
• Trading of gold, salt and cloth were also booming in Africa thanks to the Trans-Saharan
Trade Route from North Africa to West Africa. These trade routes first emerged in the 4th
century AD and by the 11th century AD caravans of over a thousand camels would carry
goods across the Sahara desert.
• Currency Used: Starting from metal in the shape of small knives and spades made of
bronze to first manufactured coins in India and China, minted coins to bills of exchange.
•
History of Money
Evolution of the Stock Exchange
1531: Belgium boasted a stock exchange in Antwerp. Brokers
and money lenders would meet there to deal in business,
government and even individual debt issues. There were no
promissory notes and bonds or real stocks.
East India House in Leadenhall Street,
London
1600s: East India companies formed which changed the way
business was done. The stock of these companies would
pay dividends on all the proceeds from all the voyages the
companies undertook. These were the first modern joint stock
companies. This allowed the companies to demand more for their
shares and build larger fleets. The profit for investors was based
on the size of the companies, combined with royal charters
forbidding competition.
America’s…
• 1773: The first stock exchange in London was officially formed
• 1780s: Japanese trader invents candlestick patterns to predict
price movements in the rice market. 1790: The birth of U.S.
investment markets was marked with the federal government
issuing $80 million in bonds to repay Revolutionary War debt.
Two years later the “Buttonwood Agreement” was signed by 24
stockbrokers and later moved to the Tontine Coffee House for
trade.
• 1792: NYSE acquires first traded securities
• 1817: The constitution of NYSE is adopted
India….
• 1830s: For the first time trading on corporate stocks and shares in
Bank and cotton presses took place in Bombay.
• 1856: An informal group of 22 stockbrokers with a then princely
amount of Rupee 1 started investing under a banyan tree opposite
the Town Hall of Bombay from the mid-1850s. The same banyan tree
still stands in the Horniman Circle Park in Mumbai.
• 1860: The exchange flourished with 60 brokers. In fact, the 'Share
Mania' in India began when the American Civil War broke and the
cotton supply from the US to Europe stopped. Further, the brokers
increased to 250. An informal group of stockbrokers organized
themselves as the “The Native Share and Stockbrokers Association”
which in 1875, was formally organized as the Bombay Stock
Exchange (BSE). Pav Bhaji is a side culinary contribution as per
folklore
1875: BSE, was set up in the year 1875 and is the oldest stock
exchange in Asia. It has evolved into its present status as the
premier stock exchange.
Back to Wall Street
• 1840s: During the California Gold Rush, curbstone brokers generated market
opportunities for mining companies, facilitating the development of a new and
rapidly growing industry.
• 1859: Oil stocks traded on the curb market after the discovery of Petroleum in
western Pennsylvania.
• 1864: Founded in part by former curbstone brokers, Open Board of Stock
Brokers started functioning. It got merged with the New York Stock Exchange in
1869.
• 1904: Emanuel S. Mendels started to organize the curb market to encourage
sound and ethical dealings. The New York Curb Market Agency was established
in 1908 to codify trading practices.
• 1921: The New York Curb Market found a new home indoors to a new building on
Greenwich Street, Lower Manhattan.
• 1944: The New York Curb Market was formed with a constitution that sets higher
brokerage and listing standards.
Digital Revolution..
The online stock trading accounts did not exist in the 1960s. The
order booking process was initiated by calling your broker and
asking him to enter the order in his own system, on your behalf.
In case the stock that was ordered is traded in the NYSE, it was
called onto the floor of the NYSE e.g. order to buy 100 shares
was matched with an order to sell 100 shares in the broker’s
system.
• If it was an over-the-counter (OTC) stock, that is, one not listed
on the NYSE or AmEx but still traded, the broker called around
by phone to market makers who quoted different prices to buy
or sell the stock.
Digital Revolution…contd
• 1969: In order to allow brokers to post offers to buy and sell
stocks after regular market hours, Instinet is founded as the first
Electronic Communication Network (ECN)
• 1970s: Founding of NASDAQ.
• 1971: The National Association of Securities Dealers, an
association of over-the counter (OTC) market makers formed in
1939, created the first electronic stock market: the National
Association of Securities Dealers Automated Quotations
(NASDAQ) market.
Digital Revolution….contd…
• 1975: Fixed commissions are abolished by the SEC. This
allowed for the rise of discounted commissions and facilitated
the growth of Charles Schwab and others. The Amex launches
its options market.
• 1976: NYSE introduces its Designated Order Turnaround (DOT)
system, which allowed brokers to route 100-share order directly
to specialists on the floor. These were not true electronic
executions since the specialist still matched the orders, but it
did bypass floor brokers.
• 1980s: The rise of electronic trading.
Digital Revolution….
• 1984: NYSE adopts a more sophisticated SuperDOT system
that allows orders up to 100,000 shares to be routed directly to
the floor. More floor brokers cut out.
• 1987: The 1987 crash is blamed partly on "portfolio insurance"
(shorting stock index futures against a stock portfolio).
Electronic trading takes another leap forward as NASDAQ
expands the Small Order Execution System (SOES), which
allows dealers with small trades to enter their orders
electronically rather than over the phone. This was done
because during the 1987 crash many broker-dealers simply
stopped answering their phones.
• 1990-95: the rise of online trading.
Back in India-Dalal Street…
• 1956: The Government of India recognized the Bombay Stock
Exchange as the first stock exchange in the country under the
Securities Contracts (Regulation) Act.
• 1994: NSE started trading on 4 November 1994. Within less
than a year, NSE turnover exceeded the BSE.
• 1996-1999: Online trading begins to explode as Internet traffic
dramatically increases. Small traders suddenly had the same
access to real-time pricing as professional brokers. The word
"day trader" enters the vocabulary.
As Data is accumulated we enter the era of
decimalization, Algo and HFT
• 2000: The NASD spins off NASDAQ into a publicly traded company.
• 2001: Stock trading in pennies begins. NYSE introduced Direct+, which facilitated
immediate automatic execution of limit orders up to 1099 shares. This is real electronic
trading (automated matching of buy and sell orders) and was the beginning of the end of
the old floor specialist system.
• 2005: Reg-NMS changes everything- HFT goes prime time. The SEC consolidates all
rules on the national market system into Reg-NMS, which forced the NYSE to go
electronic and fostered the growth of competing ECNs and exchanges.
• NYSE launches NYSE Hybrid in December, which attempts to combine the NYSE floor
operations with electronic trading occurring off the floor. The 1,099 share limits on the
Direct+ system were removed. Specialist participation in the marketplace started to drop
drastically.
• 2006: NYSE demutualizes, becomes a for-profit and publicly traded company. This gives
• 2007: NYSE merges with Euronext, which had been formed in 2000 through the merger
of the Amsterdam, Brussels, and Paris exchanges.
• 2008: NYSE eliminates specialists, renaming them Designated Market Makers, though
still in charge of maintaining a fair and orderly market in their stock

• Algorithimic trading started in India in 2008

• 2009: Credit Suisse’s Advanced Execution Services (AES) unit launched
algorithmic trading in Indian equities. The AES suite of algorithms included
traditional algorithmic strategies that seek to divide trading volumes up
over time and strategies that seek to trade at the Volume Weighted
Average Price of a stock.
• The Same year, a software developer Satoshi Nakamoto proposed bitcoin,
which was an electronic payment system based on mathematical proof.
• 2010: Direct Edge, formerly an ECN, becomes an exchange. In India,
National Stock Exchange (NSE) started offering additional 54 co-location
server ‘racks’ on lease to broking firms from June 2010 in an effort to
improve the speed in trading. Also, adapts FIX protocol. Multiple
QuantInsti, Asia’s pioneer in Algo trading education is launched and
started first algo trading education programme in India.
• 2012: NYSE created something called a single-stock circuit breaker. If the
Dow drops by a specific number of points in a specific period of time, then
the circuit breaker will automatically halt trading. This system is designed
to reduce the likelihood of a stock market crash and, when a crash occurs,
limit the damage of a crash.
• 2013: About 70% of US equities in 2013 were accounted for by
Automated Trading. Algo trading accounted for a third of the
total volume on Indian cash shares and almost half of the
volume in the derivatives segment.
• 2015: Social media integration; Bloomberg Terminals
incorporates live Tweets into its economic data service.
Bloomberg Social Velocity tracks abnormal spiked in chatter
about specific companies.
• 2017: Nasdaq smashes 6,000 and world stocks hit new high.
•.
What led to Algorithmic Trading?
Exchanges

Algo Trading

law Digital Data

Limitations
Cannot apply for
-Trading where data is not available
-No formal exchange
-Illegal trading
Through the course we will assume that all discussions are on trading
taking place through recognized exchanges
Exchanges
While we will get into detail we will look at
-Stock Exchanges-BSE/NSE
-Commodities Exchange
-Energy Exchange
- Metal Excahnge
- Currency Exchange
We will look at the Stock Exchange as an exemplar for trading examples
Class 2-Exchanges &
Products
The reference book

• Goes into great depth on exchanges

• Not so much on trading
• Not heavy on technology
A brief overview

• Trading is a search problem. Buyers must find sellers, and sellers must find buyers. Every trader wants to
trade at a good price. Sellers seek buyers willing to pay high prices. Buyers seek sellers willing to sell at low
prices. Traders also must find traders who are willing to trade the quantities, or sizes, they desire. Traders who
want to trade large quantities may have to find many willing traders to complete their trades
• Dealers and brokers help people trade. Dealers trade with their clients when their clients want to trade. The
prices at which a dealer will buy and sell are the dealers’ bid and ask prices. After they trade with their clients,
dealers then try to trade out at a profit by selling what they have bought or by buying back what they have
sold. In effect, clients pay dealers to take their trading problems. The dealers then try to solve them at a profit.
Dealers profit by buying low and selling high. Successful dealers must be excellent traders.
• Brokers are agents that arrange trades for their clients. They help their clients find traders who are willing to
trade with them. They profit by charging commissions
• In the digital age Dealers are possibly rare. IB does the role of a broker- example CCD-VGS-LnT
 A brief overview…contd
• Patient traders obtain better prices than impatient traders do because they are willing to search
longer and harder to arrange their trades at favorable terms. Impatient traders pay for the privilege
of trading when they want to trade
• Traders who offer to trade give other people options to trade. These options sometimes are quite valuable.
Traders who expose their offers can lose to clever traders who use various front- running trading strategies to
extract these option values. Traders therefore must expose their offers very carefully. They should expose
only to traders who are most likely to trade with them.
• Traders who trade only to accommodate other traders risk trading with, and losing to, well- informed
speculators. Speculators are traders who trade to profit from information they have about future prices. Well-
informed speculators can predict futures prices better than other traders can. They then choose to buy or sell
based upon which side they expect will be profitable.
• Dealers lose to well-informed speculators because they end up being on the wrong side of the trade. Prices
tend to move against their positions before they can trade out of them. All traders try to avoid trading with
well-informed speculators or copy them 
A brief overview…contd
Dealers recover their losses to informed speculators by widening the spread between the bid and ask prices at
which they will buy and sell. Uninformed traders therefore pay more for their trades when dealers lose a lot to
informed traders. In effect, uninformed traders lose to well- informed traders through the intermediation of
dealers. This why dealers are now almost extinct in some markets due to digital intervention
Traders who can estimate fundamental values cause prices to reflect their value estimates. They buy when price
is below their value estimates and sell when price is above. Their buying pushes prices up, and their selling pulls
prices down. They do not trade if they believe that prices reflect values. Well-informed traders make prices
informative.
Bluffers can sometimes fool uninformed traders into trading unwisely. In general, they can profit if the price
impacts of their buying and selling are not exactly opposite to each other
Contd…
• Since dealers may trade when bluffers want them to trade, dealers must be highly disciplined to avoid
losing to bluffers.
• Trading is a zero-sum game when gains and losses are measured relative to the market average. In a zero-
sum game, someone can win only if somebody else loses. On average, well-informed speculators and
bluffers win, and poorly informed traders and foolish traders lose. Informed traders can only profit to the
extent that less informed traders are willing to lose to them.
• Poorly informed traders trade for many reasons. Investors use the markets to move money from the present
to the future. Borrowers do the opposite. Hedgers trade to manage financial risks they face. Asset
exchangers trade one asset for another they value more. Gamblers trade to entertain themselves.
Exchanges and brokerages design markets to minimize the search costs
of trading. They usually organize markets so that everyone who wants to
trade gathers at the same place. A common gathering place helps traders
find those traders who will offer the best prices
Principles
• Exchanges and brokerages once exclusively organized their markets on physical trading floors. Now they
can organize their markets within computerized communications networks that allow buyers and sellers to
arrange their trades remotely. Electronic marketplaces have rapidly expanded as the costs of electronic
communications technologies have dropped.
• Most traders want to trade in well-established markets because other traders trade there. When many
traders trade in the same place, arranging trades is very easy. The attraction of traders to other traders
makes it hard to start new markets.
• Entrepreneurs create new markets when old markets do not adequately meet the needs of a significant set of
traders. Since traders face a diversity of trading problems, no single market can best meet every trader’s
needs. Many diverse markets may form as exchanges and brokerages compete to attract traders.
• Arbitrageurs ensure that prices do not vary much across markets. When prices diverge, they buy in cheaper
markets and sell in more expensive markets. The effect of their trading is to connect sellers in cheaper
markets to buyers in more expensive markets.
An exchange’s trading rules affect the quality of its markets. They
determine the balance of power between informed traders and
uninformed traders, between public traders and professional traders, and
between large traders and small traders. Trading rules are very
important.
Markets work best when they trade fungible instruments. An instrument is fungible
if one unit (a share, a bond, a contract, …) of the instrument is economically
indistinguishable from all other units. If so, buyers do not care which units they
receive. Since all sellers offer identical units, buyers can buy from any seller who
offers an attractive price. Sellers likewise can sell to any buyer. Fungible
instruments therefore are easier to trade than are instruments that have idiosyncratic
characteristics. In derivative markets, the benefits of fungible instruments cause
trading to concentrate in just a few standardized contracts eg UTI, Mutual funds
WHAT ARE THE DRIVERS OF A GOOD
EXCHANGE?
Information asymmetries

Traders who know more about values and traders who know more about what other
traders intend to do have a great advantage over those who do not. Well- informed
traders profit at the expense of less informed traders. Less informed traders therefore
try to avoid well-informed traders. Pay attention to who is well informed and to how
traders learn about values.
Options.

The option to trade is valuable. People who write limit orders give free
trading options to other traders. Clever traders can extract the value of
these options. Pay attention to when traders create trading options and to
how they prevent other traders from extracting their values.
Externalities
People create positive externalities when they do something that
benefits other people without compensation. People create negative
externalities when they do something that harms other people without
penalty. The most important externality in market microstructure is the
order flow externality. Traders who offer to trade give other traders
valuable options to trade for which they are not compensated. The order
flow externality attracts and binds traders to markets because traders
want to benefit from free trading options. Pay close attention to when,
why, and how traders offer to trade. Pay attention also to how markets,
brokerages, and dealers benefit from the order flow externality.
Market Structure
• Market structure consists of the trading rules, the physical layout, the information presentation systems, and
the information communication systems of a market. Market structure determines what traders can do, and
what they can know. It therefore affects trader strategies, the power relationships among different types of
traders, and ultimately trader profitability. Always consider what effects market structures have on trading
strategies and on the balance of power between various types of traders.
Competition with free entry and exit

Traders compete in markets to make profits. Trading strategies that generate large
profits attract traders who want to participate in those profits. Their entry lowers the
profits that everyone makes, on average. Conversely, traders quit using trading
strategies that are not profitable, which allows remaining traders to make more
profits on average. Free entry and exit ensures that alternative trading strategies
produce equal net profits, on average, after accounting for all costs. Wherever you
see people competing, consider how the costs of entry and exit affect their ability to
maintain profits or avoid losses. This principle will help you understand the
determinants of bid/ask spreads, dealer profits, informed trader profits, and order
submission strategies.
Communications and computing technologies

Markets are essentially information processing mechanisms. They process

information about who wants to trade, how much, and at what prices. The resulting
prices aggregate information about fundamental values. The growth in information
technologies has changed and will continue to change how people trade. Pay
attention to the role of information processing technologies in the markets.
Price correlations
Markets for similar instruments are closely related. They tend to have
similar market conditions, and they often compete fiercely with each
other for order flow. The order flow externality generally ensures that
one market among a set of closely related markets will eventually
dominate the others. Pay attention to markets that trade similar
instruments and to the differences among them that make them unique.
These issues affect how markets compete with each other.
Principal-agent problems

Principal-agent problems arise when agents do what they want to do

rather than what their principals want them to do. The most important
principal-agent problem in market microstructure involves brokers and
their clients. Brokers do not always do what you want them to do, and
they may not work as hard on your behalf as you would. Pay close
attention to how traders control their brokers.
Trustworthiness and creditworthiness

People are trustworthy if they try to do what they say they will do.
People are creditworthy if they can do what they say they will do. Since
people often will not or cannot do what they promise, market
institutions must be designed to effectively and inexpensively enforce
contracts. Pay close attention to the mechanisms that ensure that traders
will settle their trades. Attempts to solve trustworthiness and
creditworthiness problems explain much of the structure of market
institutions.
The zero-sum game

All trades involve two or more parties. The accounting gains made by
one side must equal the accounting losses suffered by the other side.
Understanding the origins of trading profits therefore requires that we
understand both sides of a trade. We must understand why traders on
one side expect to profit, and why traders on other side either are willing
to lose or do not understand that they should expect to lose
Bulls and Bears

• Traders call rising markets bull markets and falling markets bear markets. According to legend, these terms
originated from morbid contests that promoters once staged between bulls and bears. Bulls fight by thrusting
upward with their horns. In contrast, bears fight by striking downward with their claws. This image has
generated a small cottage industry of artisans who create bull- fighting-bear sculptures that traders buy to
adorn their offices and living rooms.
Who Are the Players?
Traders
• Traders are people who trade. They may arrange their own trades, they may have others arrange trades for
them, or they may arrange trades for others. Proprietary traders trade for their own accounts while brokers
arrange trades as agents for their clients. Brokers are also called agency traders, commission traders, or
commission merchants. Proprietary traders engage in proprietary trading, and brokers engage in agency
trading
• Traders have long positions when they own something. Traders with long positions profit when prices rise.
They try to buy low and sell high.
• Traders have short positions when they have sold something that they do not own. Traders with short
positions hope that prices will fall so that they can repurchase at a lower price. When they repurchase, they
cover their positions. Short sellers profit when they sell high and buy low
Buy and Sell Side and Liquidity
• The trading industry has a buy side and a sell side. The buy side consists of traders who buy exchange
services. Liquidity is the most important of these services. Liquidity is the ability to trade when you want to
trade. Traders on the sell side sell liquidity to the buy side. A substantial fraction of this book considers how
interactions between buy side and sell side traders determine the price of liquidity.
• (The buy- and sell sides of the trading industry have nothing to do with whether a trader is a buyer or seller
of an instrument. Traders on both sides of the trading industry regularly buy and sell securities and
contracts. The terms buy side and sell side refer to buyers and sellers of exchange services.)
Buy Side
• The buy side of the trading industry includes individuals, funds, firms, and governments that use the
markets to help solve various problems that they face. These problems typically originate outside of trading
markets. For example, investors use securities markets to solve intertemporal cash flow problems: They
have income today that they would like to have available in the future. They use the markets to buy stocks
and bonds to move their income from the present to
•
the future. We discuss this problem and other buy side trading problems in Chapter 8 (Why People Trade).
• Many buy side institutions are pension funds, mutual funds, trusts, endowments, and foundations that invest
money. These institutions are known collectively as investment sponsors. Investment sponsors frequently
employ investment advisors to manage their funds. Investment advisors are also called investment counselors,
investment managers, or portfolio managers. Investment advisors often employ traders to implement their
trading decisions. These traders are buy side traders. The people and institutions who will ultimately benefit
from the funds that investment sponsors hold are beneficiaries.
Trader type Generic examples Why they trade Typical instruments
Investors Individuals To move wealth from the present to Stocks Bonds
Corporate pension funds Insurance the future for themselves or for their
funds Charitable and legal trusts clients.
Endowments
Mutual funds Money managers

Borrowers Homeowners To move wealth from the Mortgages Bonds Notes

Students Corporations
future to the present
Hedgers Farmers Manufacturers Miners To reduce business Futures contracts Forward
Shippers
Financial institutions operating risk. contracts Swaps

Asset exchangers International corporations To acquire an asset that Currencies Commodities

Manufacturers they value more than the
Travelers asset that they tender

Gamblers Individuals To entertain themselves Various

Sell Side
• The sell side of the trading industry includes dealers and brokers who provide exchange services to the buy
side. Both types of traders help buy side traders trade when they want to trade.
• Dealers accommodate trades that their clients want to make by trading with them when their clients want
to trade. Dealers profit when they buy low and sell high. We discuss dealers in Chapter 13.
• In contrast, brokers trade on behalf of their clients. Brokers arrange trades that their clients want to make by
finding other traders who will trade with their clients. Brokers profit when their clients pay them
commissions for arranging trades with other traders.
• Many sell side firms employ traders who both deal and broker trades. These firms therefore are known as
broker-dealers or dual traders
• The sell side exists only because the buy side will pay for its services. We therefore must understand why
the buy side trades before we can understand when the sell side is profitable
Trader type Generic examples Well-known US examples Why they trade
Dealers Market makers Spear Leads & Kellogg LaBranche & To earn trading profits by
Co., Inc. Bernard L. Madoff Investment
Specialists Floor traders Locals supplying liquidity.
Securities
Day traders Scalpers
Knight Trading Group TimberHill LLC

Brokers Retail brokers Discount brokers Full- Charles Schwab & Co. E*Trade To earn commissions by
service brokers Institutional brokers Dreyfus Brokerage Services Abel/Noser
Block brokers Futures commission Corp. arranging trades for
merchants XpressTrade clients.
Cargill Financial Markets Group

Broker-dealers Wirehouses Goldman Sachs Merrill Lynch Salomon To earn trading profits
Smith Barney
Morgan Stanley Dean Witter Credit and trading commissions.
Suisse First Boston
What is a Wirehouse
Traders often call large broker-dealers wirehouses. The word “wire” in wirehouse once referred to the telegraph.
Following its invention, broker-dealers used the telegraph to collect orders from branch offices in distant cities.
Those who quickly adopted it were able to expand their businesses substantially and thereby greatly increase
their profits. The ability to communicate quickly was—and remains—very important in the trading industry
Trade Facilitators
• Depository
• Custodians
Orders and Order Properties
• Orders are trade instructions. Orders specify what traders want to trade, whether to buy or sell, how much,
when and how to trade, and most importantly, on what terms. Traders issue orders when they cannot
personally negotiate their trades.
• Orders are the fundamental building blocks of trading strategies. To trade effectively, you must specify
exactly what you intend. Your order submission strategy is the most important determinant of your success
as a trader. The proper order used at the right time can make the difference between a good trade, a costly
trade, or no trade at all.
• Many markets arrange all their trades by using a set of rules to match buy and sell orders that traders submit
to them. To understand how these markets work and to use them effectively, you must understand how
traders specify their orders.
• Understanding orders will also allow you to see where liquidity comes from. Liquidity is the ability to trade
when you want to trade. Some orders offer liquidity by presenting other traders with trading opportunities.
Other orders take liquidity by seizing those opportunities. Trader decisions to offer or take liquidity
therefore affect market quality. To understand liquidity, you must understand how traders form their order
submission strategies.
• Traders choose orders that have properties that allow them to best solve their trading problems.
• Traders use specialized words and phrases to communicate quickly and accurately with each other. Whether
you intend to trade or simply want to learn about trading, you need to be familiar with market nomenclature
• Although order instructions have the same meanings in all markets, their properties differ depending on the
type of market to which traders submit them
• Mostlytraders submit their orders to a continuous trading market that arranges trades as orders arrive.
Identical orders have slightly different properties in call markets that collect and process all orders at the same
time
What Are Orders, and Why Do People Use Them?

• Orders are instructions that traders give to the brokers and exchanges that arrange their trades. The
instructions explain how they want their trades to be arranged.
• An order always specifies which instrument (or instruments) to trade, how much to trade, and whether to buy
or sell. An order may also include conditions that a trade must satisfy. The most common conditions limit the
prices that the trader will accept. Other conditions may specify for how long the order is valid, when the order
can be executed, whether it is okay to partially fill the order, where to present the order, and how to search for
the other side. Some orders even specify the traders with whom the trader is willing to trade
• Orders are necessary because most traders do not personally arrange their trades. Traders who arrange their
own trades—typically dealers—do not use orders. They decide on the spot what they want to do and how to
do it. All other traders must carefully express their intentions ahead of time.
• For many small traders, it is not economical to continuously monitor the market. These traders use orders to
represent their interests when they are not paying close attention to the market.
• Traders who arrange their own trades have an advantage over traders who use orders to express their
intentions. The former can respond to market conditions as they change. The latter must anticipate such
changes and write contingencies into their orders to deal with them. Carefully written orders will adequately
represent traders’ interests even when conditions change. When orders do not do so, traders must cancel them
and resubmit new instructions. During the time it takes to cancel and resubmit orders, traders can lose because
their old orders trade before they can cancel them, or because they cannot submit new orders in time to take
advantage of the changing market conditions. Traders therefore must carefully specify their intentions when
they use orders to trade.
• In general, traders who can respond most quickly to changes in market conditions have an advantage over
slower traders. Traders who submit and cancel orders manually are slower than traders who use computers
to monitor and adjust their orders. Where speed is of the essence, floor traders and computerized traders are
the most successful traders.
• Clear and efficient communication is essential when trading in fast markets. Brokers must understand
exactly what traders intend. Otherwise, extremely costly errors may occur. To avoid mistakes, most traders
use standard orders to decrease the probability that they will misunderstand each other when communicating
quickly. All traders recognize and understand these orders.
Class3-03-03-2021
Exchanges and Products
John C. Hull is a Professor of Derivatives and Risk Management at
the Rotman School of Management at the University of Toronto.[He is a
respected researcher in the academic field of quantitative finance (see for
example the Hull-White model) and is the author of two books on financial
derivatives
that are widely used texts for market practitioners: "Options, Futures, and
Other Derivatives"[5] and "Fundamentals of Futures and Options
Markets".[6] He has also written "Risk Management and Financial
Institutions" and "Machine Learning in Business: An Introduction to the
World of Data Science"He studied Mathematics at Cambridge
University (B.A. & M.A.), and holds an M.A. in Operational
Research from Lancaster University and a Ph.D. in Finance from Cranfield
University. In 1999, he was awarded the Financial Engineer of the Year
Award, bythe International Association of Financial Engineers. He has also
won many teaching awards, such as the University of Toronto's prestigious
Northrop Frye award.[7]
Professor Basu’s areas of interest in terms of research are in Financial
Calculus, Option Pricing, Bond and Portfolio Valuation, Applications of
Quantitative Techniques to Finance, Insurance, Reinsurance, Risk
Management, Biostatistics and Corporate Finance. He has presented his
work at international conferences in Thailand, USA, UK, France, Spain,
Norway, Iran, Tunisia, Morocco, The Netherlands and India. He has, to his
credit, a number of papers in international journals of repute like Insurance:
Mathematics and Economics and the Journal of Statistical Planning and
Inference. He is currently President of the Asia-Pacific Risk and Insurance
Association (APRIA)..
One view of Buy Side or Sell Side
Some Reading material….
• Belonsky Gail M., and David M. Modest, 1993, Market microstructure: An empirical retrospective,
Working paper, Haas School of Business, University of California, Berkeley.
• Coughenour, Jay, and Kuldeep Shastri, 1999, Symposium on market microstructure: A review of
empirical research, Financial Review 34(4), 1-28
• Dalton, John M., ed., 1993. How the Stock Market Works (New York Institute of Finance, New York,
NY).
• Downes, John, and Jordan E. Goodman, eds., 1991. Dictionary of Finance and Investment Terms
• (Barron’s Educational Series, New York, NY).
• Fan, Ming, Sayee Srinivasan, Jan Stallaert, and Andrew B. Whinston, 2002. Electronic Commerce and
the Revolution in Financial Markets (South-Western, Mason, OH).
• Kalman J. Cohen, Steven F. Maier, Robert A. Schwartz, and David K. Whitcomb, 1986. The
Microstructure of Securities Markets (Prentice-Hall, Englewood Cliffs, NJ).
• Lee, Rubin, 1998. What is an Exchange: The Automation, Management and Regulation of Financial
Markets (Oxford University Press, New York, NY).
• Lyons, Richard K., 2002. The Microstructure Approach to Exchange Rates (The MIT Press,
Cambridge, MA).
•
Some Reading material
• Madhavan, Ananth, 2000, Market microstructure: A survey, Journal of Financial Markets
3(3), 205-258
• O’Hara, Maureen, 1995. Market Microstructure Theory (Basil Blackwell, Cambridge, MA).
Schwartz, Robert A., 1991. Reshaping the Equity Markets: A Guide for the 1990s (Harper
• Business, New York, NY).
•
Sharp, Robert, 1989. The Lore and Legends of Wall Street (Dow Jones-Irwin, New York, NY).
• Stoll, Hans R., 1992, Principles of trading market structure, Journal of Financial Services
Research 6(1), 75-107
• Teweles, Richard J., Edward S. Bradley, and Ted M. Teweles, 1998. The Stock Market 7th ed.
(John Wiley and Sons, Inc., New York, NY).
• Teweles, Richard J., and Frank J. Jones; Ben Warwick, ed., 1999. The Futures Game: Who
Wins? Who Loses? And Why? 3rd ed. (McGraw-Hill, New York, NY).
• Wagner, Wayne, ed., 1989. The Complete Guide to Securities Transactions: Enhancing
Performance and Controlling Costs (John Wiley & Sons, New York, NY).
•
Liquidity
• Liquidity in the economy
• Liquidity of a Share
• Liquidity in a exchange . Buy-Sell?
Orders
• Orders are trade instructions. Orders specify what traders want to trade, whether to buy or sell, how much,
when and how to trade, and most importantly, on what terms. Traders issue orders when they cannot
personally negotiate their trades.
• Orders are the fundamental building blocks of trading strategies. To trade effectively, you must specify
exactly what you intend. Your order submission strategy is the most important determinant of your success
as a trader. The proper order used at the right time can make the difference between a good trade, a costly
trade, or no trade at all.
• Many markets arrange all their trades by using a set of rules to match buy and sell orders that traders submit
to them. To understand how these markets work and to use them effectively, you must understand how
traders specify their orders
• Understanding orders will also allow you to see where liquidity comes from. Liquidity is the ability to trade
when you want to trade. Some orders offer liquidity by presenting other traders with trading opportunities.
Other orders take liquidity by seizing those opportunities. Trader decisions to offer or take liquidity therefore
affect market quality. To understand liquidity, you must understand how traders form their order submission
strategiesorders are, how traders specify them, and most importantly, what properties they have.
• Traders choose orders that have properties that allow them to best solve their trading problems
The Art of Communication
• The General 6 am…
• My experience in Mynamar
• Should be precise, unambiguous and actionable
• Especially in the digital age where the parties are just “entities”
• Familiarize yourself with the many trading terms involved in your exchange
• Traders use specialized words and phrases to communicate quickly and accurately with each other. Whether
you intend to trade or simply want to learn about trading, you need to be familiar with market nomenclature.
• Although order instructions have the same meanings in all markets, their properties differ depending on the
type of market to which traders submit them
• Most markets are continuous trading markets that arrange trades as they arrive
• Identical orders have slightly different properties in call markets that collect and process all orders at the same
time
What are orders?
• Orders are instructions that traders give to the brokers and exchanges that arrange their trades. The
instructions explain how they want their trades to be arranged.
• An order always specifies which instrument (or instruments) to trade, how much to trade, and whether to buy
or sell. An order may also include conditions that a trade must satisfy. The most common conditions limit the
prices that the trader will accept. Other conditions may specify for how long the order is valid, when the order
can be executed, whether it is okay to partially fill the order, where to present the order, and how to search for
the other side. Some orders even specify the traders with whom the trader is willing to trade.
• Orders are necessary because most traders do not personally arrange their trades. Traders who arrange their
own trades—typically dealers—do not use orders. They decide on the spot what they want to do and how to
do it. All other traders must carefully express their intentions ahead of time.
• For many small traders, it is not economical to continuously monitor the market. These traders use orders to
represent their interests when they are not paying close attention to the market.
• Traders who arrange their own trades have an advantage over traders who use orders to express their
intentions. The former can respond to market conditions as they change. The latter must anticipate such
changes and write contingencies into their orders to deal with them. Carefully written orders will adequately
represent traders’ interests even when conditions change. When orders do not do so, traders must cancel them
and resubmit new instructions. During the time it takes to cancel and resubmit orders, traders can lose because
their old orders trade before they can cancel them, or because they cannot submit new orders in time to take
advantage of the changing market conditions. Traders therefore must carefully specify their intentions when
they use orders to trade.
• In general, traders who can respond most quickly to changes in market conditions have an advantage over
slower traders. Traders who submit and cancel orders manually are slower than traders who use computers to
monitor and adjust their orders. Where speed is of the essence, floor traders and computerized traders are the
most successful traders-Algo trading or quant trading
• Clear and efficient communication is essential when trading in fast markets. Brokers must understand exactly
what traders intend. Otherwise, extremely costly errors may occur. To avoid mistakes, most traders use
standard orders to decrease the probability that they will misunderstand each other when communicating
quickly. All traders recognize and understand these orders.
Terms
• Traders indicate that they are willing to buy or sell by making bids and offers. Traders quote their bids and
offers when they arrange their own trades. Otherwise, they use orders to convey their bids and offers to the
brokers or automated trading systems that arrange their trades. Bids and offers usually include information
about the prices and quantities that traders will accept. Traders call these prices bid and offer prices. They also
use the terms bidding price, offering price, asking price, or simply bid and ask. They refer to the quantities as
sizes
• Prices are firm when traders can demand to trade at those prices. Prices are soft if the traders who offer them
can revise them before trading. Orders generally have firm prices
The highest bid price in a market is the best bid. The lowest offer price is the best
offer (or equivalently, the best ask). Traders also call them the market bid and the
market offer (or market ask) because they are the best prices available in the market.
A market quotation reports the best bid and best offer in a market. A market
quotation is often called the BBO, which is the acronym for Best Bid and Offer.
Many markets continuously publicize their market quotations. The best bid and
offer anywhere in the United States is the NBBO—National Best Bid and Offer.
The difference between the best ask and the best bid is the bid/ask
spread. Traders sometimes call it the inside spread because the space
between the highest bid price and the lowest ask price is inside the
market. The English often refer to the spread as the touch. In sports
betting markets, bettors and bookies call it the vigorish.
An order offers liquidity—or equivalently supplies liquidity—if it gives
other traders an opportunity to trade. For example, suppose Joe issues an
order to buy 100 shares of IBM for no more than 100 dollars per share
from the first person to contact him before trading closes today. Joe’s
bid offers liquidity because other traders now have the opportunity to
sell IBM for 100 dollars per share. Joe’s bid is a day limit order because
it is only valid for the day, and because Joe limits the price that he will
pay
• Buyers and sellers can both offer liquidity. Buyers
offer liquidity when their bids give other traders
opportunities to sell. Sellers offer liquidity when their
offers give other traders opportunities to buy
The dual use of the word “offer” may seem confusing.
It may refer to an offer of an item for sale, or to an offer
of liquidity. If you think of liquidity—the ability to
trade when you want to trade—as a service that you can
buy or sell, the use of the word “offer” makes sense.
This perspective leads to many useful insights. For
example, dealers make money by selling liquidity to
their clients.
Standing orders are open offers to trade. Joe’s order will stand
until someone sells to Joe at 100 dollars or less, the order
expires at the end of the day, or Joe cancels it. Standing orders
are also called open orders. Since standing orders allow other
traders to trade when they want to trade, traders offer liquidity
when they have orders outstanding.
• Traders who want to trade quickly demand liquidity. Traders take liquidity
when they accept offers—standing limit orders or quotes—that other traders
made. If Sue is willing to sell 100 shares of IBM at 100 dollars, she can initiate
a trade by taking Joe’s offer.
• Traders who demand to trade immediately demand immediacy. Immediacy is
one of several dimensions of liquidity.
• A market is liquid when traders can trade without significant adverse effect on
price. Markets with many standing limit orders and small bid/ask spreads are
usually quite liquid
• The prices at which orders fill are trade prices. Buy orders that trade
at high prices and sell orders that trade at low prices trade at inferior
prices.
• Markets and traders sometimes treat orders differently depending on
whether they are agency orders or proprietary orders. Agency orders
are orders that brokers represent as agents for their clients.
Proprietary orders are orders that traders represent for their own
accounts. In many organized markets, agency orders have precedence
over proprietary orders at the same price. The tension always exists.
Trigger of a general offer
Market Structures
The trading rules and the trading systems used by a market
define its market structure. They determine who can trade;
what they can trade; and when, where, and how they can trade.
They also determine what information traders can see about
orders, quotations, and trades; when they can see it; and who
can see it.
Market structure is extremely important because it determines
what people can know and do in a market. Since power comes
from knowledge and the ability to act on it, market structure
helps determine power relations among various types of
traders. These relationships greatly affect who will trade
profitably
To trade effectively, you need to know the structure of every market in
which you trade. The trading strategies that are successful in one market
often do not work well in markets with different structures. The best
order submission strategy for a given trading problem generally depends
on the structure of the market where the trader intends to solve the
problem. Traders therefore behave differently in different markets.
You must understand market structure, and how it
affects trader behavior, to understand the origins of
market liquidity, price efficiency, volatility, and trading
profits. These variables all depend on trader behavior.
Since market structure affects trader behavior, it helps
determine whether markets will be liquid, whether
prices will be informative, and which traders will trade
profitably.
We can look at a basic framework. Keep that at the core and try to use
it for other markets. Some of th points we have already discussed
Trading Sessions
Trading takes place in trading sessions. The two types of
trading sessions are continuous trading sessions and call
market sessions. In continuous trading, traders can attempt to
arrange their trades whenever the market is open. In call
markets, all trades take place only when the market is called.
Trading forums are the places where traders arrange their
trades. In physically convened markets, traders must be on a
trading floor to negotiate their trades. Physically convened
futures markets trade in trading pits. Physically convened
stock markets trade at posts.
Today trading is electronic
Traders and exchanges use various execution systems to
arrange trades. In quote-driven systems, dealers arrange most
trades when they trade with their customers. In order-driven
systems, all trades are arranged by using order precedence
rules to match buyers to sellers and trade pricing rules to
determine the prices of the resulting trades. In brokered trading
systems, brokers arrange trades by helping buyers and sellers
find each other
Various information systems move information in and out of
the market, present it, and store it. Order routing systems send
orders from customers to brokers, from brokers to dealers,
from brokers to markets, and from markets to markets. These
systems also send reports of filled orders back to customers.
Order presentation systems present orders to traders so that
they can act upon them. The systems may use screen-based,
open-outcry, or hand-signaling technologies. Order books
store open orders. Market data systems report trades and
quotes to the public.
In most markets, traders can only use prices that are an integer
multiple of a specified minimum price increment. The size of
the increment, measured as a fraction of price, varies
considerably across markets.
Trading Sessions

Markets have trading sessions during which trades are

arranged. The two types of trading sessions are continuous
market sessions and call market sessions.
Continuous Markets

In continuous markets, traders may trade anytime while the market is

open. Trading is continuous in the sense that traders may continuously
attempt to arrange their trades. In practice, they usually trade only
when a trader demands liquidity.
Continuous trading markets are very common. Almost all major stock,
bond, futures, options, and foreign exchange markets have continuous
trading sessions.
Call Markets

• In call markets, all traders trade at the same time when the market is called. The
market may call all securities simultaneously, or it may call the securities one at a
time, in a rotation. Markets that call in rotation may complete only one rotation
per trading session, or as many rotations as their trading hours permit. Markets
that call in rotation were once very common. Now only in the stock markets of a
few small countries call in rotation.
• Many continuous order-driven exchanges open their trading sessions with call
market auctions and then switch over to continuous trading. These markets also
use calls to restart their trading after a trading halt. Open-outcry futures
exchanges, however, start continuous trading immediately when they open.
• Call markets are used as the exclusive market mechanism for many instruments.
Most governments sell their bonds, notes, and bills in call market auctions. Some
stock markets also use calls to trade their least active securities. The Deutsche
Börse and Euronext Paris Bourse are examples of such markets.
Order Driven Markets
Order-driven markets use trading rules to arrange their trades.
These markets include oral auctions, single price auctions,
continuous electronic auctions, and crossing networks.Trading
strategies depend on market structure
Order-driven markets are quite common. Almost all of the
most important exchanges in the world are order-driven
markets. Most newly organized trading systems choose
electronic order- driven market structures
• Despite the great variation in how order-driven markets
operate, their trading rules are all quite similar. All order-
driven markets use order precedence rules to match buyers
to sellers and trade pricing rules to price the resulting trades.
• Variations in trading rules distinguish order-driven markets
from each other. The trading strategies that work best in one
market may work poorly in markets with different rules.
Traders therefore need to know how trading rules affect
optimal trading strategies.
Key preferences in order driven markets
• Time
• Price
Brokers
• Why Brokers?
Market type Trades Market structure Brokerage role
Order flow Small to medium sizes Order-driven or quote- Brokers receive orders
in seasoned securities and driven and match them with
contracts orders and quotes made
by other traders
Block Large sizes in seasoned Brokered Brokers receive an order
securities and contracts. on one side
and must search for
traders who will take the
other side. Brokers
occasionally identify both
sides.

New and seasoned Large size offered by an Brokered Brokers sell securities to
offerings issuer or one or more buyers on
large holders behalf of issuers and large
holders
Mergers and acquisitions Company to company Brokered Brokers find one or both
parties
Clearing & Settlement
• Brokers used to do that
• Clearing House
• Today’s electronic exchanges do it automatically
• Harshad Mehta scam- Sitaram?
Why do people trade?
Who is he?

Nudge…..
• You must understand why people trade to use markets effectively.
Markets provide many valuable opportunities. To take advantage of
them, you must first recognize them.
• By considering why people trade, you will better understand why
you trade and whether you should trade. Many traders do not fully
recognize the reasons why they trade. Consequently, either they
pursue inappropriate trading strategies, or they trade when trading is
counterproductive to their true interests. The optimal trading strategy
for a given trading problem depends on the problem. You cannot
trade well if you do not know why you want to trade.
• Knowing why people trade may also help you determine whether other traders
understand why they are trading. This skill is very important because you can
usually distinguish a good money manager from a poor one by whether they
understand well why they trade. It is also important because traders who do not
fully understand why they trade often trade foolishly. If you can identify such
traders, you may be able to profit from their foolishness.
• If you engage in any trading strategy that depends on the volume of trade, you
must understand why people trade to interpret volumes properly. Many factors
cause people to trade. If your trading strategy depends on one of these factors,
you will want to examine volumes carefully. However, you must be careful to
recognize when other factors may cause people to trade.
• Otherwise, you may misinterpret volumes and trade when you should not.
• Markets are successful only when people trade in them. If you want to design new
markets, or if your business depends on trading in a successful market, you must
understand why—and how— people trade.
• Trading is a zero-sum game in an important accounting sense. In a zero-sum
game, the total gains of the winners are exactly equal to the total losses of the
losers. Trading is a zero-sum game because the combined gains and losses of
buyers and sellers always sum to zero. If a buyer profits from a trade, the seller
loses the opportunity to profit by the same amount.
• Likewise, if a buyer loses from a trade, the seller avoids an identical loss.
• Successful traders must understand the implications of the zero-sum game. To
trade profitably, traders must trade with people who will lose. Profit-motivated
traders therefore must understand why losers trade to know when they should
trade.
• Finally, you must understand why people trade to form well-reasoned opinions
about market structures. Different structures favor different trader types. If you
intend to influence a decision about market structure, you should consider first
how the decision affects various traders. The benefits that traders obtain from
markets depend on why they trade. Regulators and other interested parties must
therefore understand these reasons
• In practice, traders often trade for many reasons. The complexity of their motives
explains why many traders get confused and fail to fully recognize why they trade.
By considering stylized traders, we simplify our discussions and ultimately make
it easier for you to identify the different reasons why people trade.
• Profit motivated traders
• Utilitarian traders
• Informed traders
• Uninformed traders
Informed traders-Value based investing
What constitutes a good market?
Some Questions….
 Should regulators consolidate all orders into a central limit order book?
 Should markets use quote-driven or order-driven systems?
 Should regulators allow internalization and preferencing?
 Should regulators impose price limits or trading halts on trading?
?
 Should dealers yield to their customers?
 Should regulators require that markets be linked electronically? How fast
should those links be?
 What trading hours should markets adopt?
 Who should be able to see the limit order book?
 Who owns market data?
 What securities and contracts should regulators allow exchanges to trade?
The Stake holders…
 Legislators pass laws that dictate structures.
 Regulators interpret those laws, propose new ones, and selectively enforce them.
 Government administrators propose laws, veto laws, and use their influence in a myriad of ways to
promote their interests. In some countries, they also write the laws.
 Judges interpret laws and write new case law.
 Exchanges, brokers, clearing agencies, and information providers freely create any market structures that
the legal system permits. They also frequently propose—and sometimes even implement—structures that
laws and regulations do not currently permit.
 Issuers influence market structure through the decisions they make about where to list their securities.
 Traders likewise influence market structure through the decisions they make about where to trade.
 Investors and the general public influence market structure by voting for politicians that favor their
interests and by lobbying those politicians.
• Finally, the leaders of trade organizations, public interest groups, and watchdog agencies often lobby on behalf
of their constituents
Fundamentals
• Informed traders are speculators who acquire and act on information about fundamental values. They buy
when prices are below their estimates of fundamental value and sell when prices are above their estimates.
Informed traders include value traders, news traders, information- oriented technical traders, and
arbitrageurs
• A price is informative when it is near its corresponding fundamental value. Informative prices are extremely
valuable to the economy because they help us allocate resources efficiently. To fully appreciate how market-
orientated economies work, you must understand how informed traders make prices informative.
• The market value of an instrument is the price at which traders can buy or sell the instrument. The
fundamental value (or intrinsic value) is the “true value” of the instrument. In financial terms, fundamental
value is the expected present value of all present and future benefits and costs associated with holding the
instrument. Everyone would agree upon this value if they all knew everything known about the instrument, if
they all used the proper analyses to predict and discount all uncertain future cash flows, and if they all
perceived the benefits and costs of holding the instrument equally. Since these conditions never occur, traders
often differ in their opinions about fundamental values.
• Fundamental values are not perfect foresight values. Fundamental values depend only on information that
is currently available to traders. Perfect foresight values depend on all current
and future information about values. Fundamental values are the best estimates of perfect foresight values.
Noise
• Prices are completely informative when they equal fundamental values. Efficient markets produce prices that
are very informative. The difference between fundamental value and market value (price) is noise. Informed
traders try to identify the noise in prices by estimating fundamental values. Since we do not observe
fundamental values, we cannot easily determine whether prices are informative or noisy.
Fischer Black on Noise

Fischer Black was a mathematician who made many seminal contributions to the
development of financial theory. Perhaps most notably, he helped develop option-
pricing theory, for which Myron Scholes and Robert Merton received the 1997
Nobel Prize in Economic Science. Had Fischer not died two years before the Prize
was awarded, he undoubtedly also would have been a Nobel laureate.
In his 1985 presidential address to the American Finance Association, Black
offered a now famous opinion about noise. He believed that we should consider
stock prices to be informative if they are between one-half and twice their
fundamental values! Most economists believe that the prices of actively traded
securities are well within these extreme bounds, but no one can know for sure.
• Source: Fischer Black, 1986, “Noise,” Journal of Finance 41(3), 529-543
• Changes in fundamental values are completely unpredictable. Since fundamental
values reflect all available information, they change only when traders learn
unexpected new fundamental information. If fundamental value changes were
predictable, current fundamental values would not fully reflect the information
upon which the predictions are based. Fundamental value changes therefore must
be unpredictable. Since prices are very close to fundamental values in efficient
markets, price changes in efficient markets are quite unpredictable.
• When traders cannot predict future price changes, statisticians say that prices
follow a random walk. Plots of random walks through time look like paths that
wander up or down at random because random walks are completely
unpredictable.
• Informed traders estimate fundamental values. They may base their estimates on private information that only
they have, or on public information that any trader can obtain. Informed traders compare their value estimates
with the corresponding market prices. They consider instruments to be undervalued if prices are less than
their estimates of fundamental value, and overvalued if prices are greater.
• Informed traders buy instruments that they believe are significantly undervalued and sell instruments that
they believe are significantly overvalued. They hope to profit when the prices of their purchases rise, and
when the prices of their sales fall. Informed traders naturally hope that these price changes will occur
quickly.
• Informed traders lose money when they estimate fundamental values poorly. When their value estimates are
wrong, they pay too much for instruments they have overvalued, and they sell toocheaply those instruments
that they have undervalued. Informed traders who consistently estimate values poorly usually quit trading
when they have lost more money than they can tolerate or when bankruptcy forces them out of the markets.
• Informed traders also can lose money even if they accurately estimate
fundamental values. This happens when prices move away from fundamental
values rather than towards them. These losses, however, tend to be short-term
losses. In the long run, prices usually revert toward their fundamental values so
that well-informed traders ultimately profit.
• Even if prices never adjust to their fundamental values, well-informed traders
who have correctly estimated values still can profit from their trades if they are
patient. When they buy an undervalued instrument, they acquire the rights of
ownership for less than their aggregate value. By holding the instrument, they will
eventually receive the benefits of these rights—typically interest, dividends,
royalties, capital repayments, or liquidating distributions—at a lower price than
they could otherwise obtain them. When they sell overvalued instruments, they
can invest the proceeds in instruments with higher expected rates of return.
• Informed traders, like all other traders, often significantly impact prices when they trade. Their
buying tends to push prices up, and their selling tends to push prices down. Since they buy when
price is below their estimates of fundamental value and sell otherwise, the effect of their trading is
to move prices toward their estimates of fundamental value. Their trading therefore causes prices
to reflect their estimates of fundamental value. When informed traders accurately estimate values,
their trading makes prices more informative.
• Informed traders generally differ in their estimates of value. This often happens when they base
their estimates on different data. Informed traders often trade with each other so that the price
impacts of their trading tend to cancel. The net impact of their trading is a market price that
reflects an average of their different value estimates. This price usually is more informative than
are any of the individual value estimates. Markets aggregate data from many sources to produce
prices that typically estimate fundamental values more accurately than can any individual trader.
Volatality
• Volatility is the tendency for prices to unexpectedly change. Prices change in
response to new information about values and in response to the demands of
impatient traders for liquidity.
• Volatility itself changes through time. Sometimes prices are very volatile. Other
times, prices are very stable and hardly change at all. Large price changes
sometimes occur in short time intervals. Regulators and traders refer to episodes
of such price changes as episodic volatility. Episodic volatility concerns many
people because it can be quite scary.
• Volatility, risk, and profit are closely related. Every drop in prices creates losses
for traders who have long positions and profits for traders with short positions.
Likewise, every price rise causes losses for traders with short positions and
profits for traders with long positions. Traders therefore are very interested in
volatility because it can have a significant impact on their wealth. If risk scares
you or profits interest you, you need to know about volatility.
• Volatility especially concerns options traders. Option contract values depend
critically on the volatility of the underlying instrument. Options traders must be
able to measure and predict volatilities to trade profitably. Both skills require that
they understand well the origins of volatility.
• Technical traders who try to interpret trading volumes also pay close attention to
volatility because volumes and volatility are often correlated. The relation
between the two variables is not simple, however. It depends on the origins of
the volatility.
• Volatility greatly concerns regulators. Excessive volatility may indicate that
markets are not functioning well. Since accurate prices are extremely important in
the economy, regulators pay close attention to the markets when prices are highly
volatile. They are especially attentive when markets crash.
• Volatility is of two types. Fundamental volatility is due to unanticipated changes
in instrument values while transitory volatility is due to trading activity by
uninformed traders.
• The distinction is important both for traders and for regulators. Traders must
distinguish between the two volatility types to accurately predict future volatility,
the profitability of dealing strategies, and transaction costs. Regulators must
distinguish between them because they cannot have any lasting effect on
fundamental volatility, but they often can substantially affect transitory volatility.
Depending on the policies regulators adopt, they may decrease or increase
transitory volatility.
• Since economies use prices to allocate resources, it is very important that prices
reflect fundamental values. Values change when the fundamental factors that
determine them change. Prices therefore should change when people learn that
fundamental factors have unexpectedly changed. Such price changes contribute
to fundamental volatility.
• When new information about changes in fundamental values is common
knowledge, prices may change without any trading. For example, suppose that an
unexpected killer frost descends upon Florida overnight. The morning news will
undoubtedly report the event. The next day, orange juice futures contracts will
open at a much higher price than the last price of the previous trading day.
• When only a few people know new information about changes in
fundamental values, prices generally will change on high trading
volumes. The well-informed traders will trade on their information.
The pressures their trades put on prices will cause prices to change
to reflect the new fundamental values.
• Since informed traders generally hurt dealers, and since dealers
generally do not know when they trade with informed traders, dealers
try to infer information about fundamental values from their order
flows.
Fundamental factors
Any factor that determines the value of a trading instrument can cause
the price of that instrument to change. For a commodity, the most
important factors are cash market supply and demand conditions. Other
important factors are interest rates and storage costs. For a bond, the
most important factors are interest rates and the credit quality of the
issuer. For a stock, the most important factors are quality of
management, the values of the company’s resources and technologies,
the supply and demand conditions in its product markets and in its input
markets, and interest rates. For currencies, the important valuation
factors include national inflation rates, macroeconomic policies, and
trade and capital flows. Unexpected changes in any of these factors
generate fundamental volatility in the instrument
Expected changes in fundamental factors generally do not change
prices. Informative prices usually fully incorporate all available
information about future values. Since people base their expectations on
existing information, fully informative prices will already incorporate
expected changes in fundamental factors. When the expected event
occurs, it is not surprising, and it therefore should not cause prices to
change. Only unexpected events cause fundamental price volatility.
Consequently, the identifying characteristic of fundamental volatility in
fully informative prices is unpredictable price changes. An
unpredictable price process is called a random walk.
Bubbles & Crashes
Assignment
• Open a demat account
• Buy and sell at least 1 share
• Write 1 page on the order information etc for buying
• Write 1 page for the order information on selling
 06032021
Class 4-Exchanges, Products
Noise
• Prices are completely informative when they equal fundamental values. Efficient markets produce prices that
are very informative. The difference between fundamental value and market value (price) is noise. Informed
traders try to identify the noise in prices by estimating fundamental values. Since we do not observe
fundamental values, we cannot easily determine whether prices are informative or noisy.
Fischer Black on Noise

Fischer Black was a mathematician who made many seminal contributions to the
development of financial theory. Perhaps most notably, he helped develop option-
pricing theory, for which Myron Scholes and Robert Merton received the 1997
Nobel Prize in Economic Science. Had Fischer not died two years before the Prize
was awarded, he undoubtedly also would have been a Nobel laureate.
In his 1985 presidential address to the American Finance Association, Black
offered a now famous opinion about noise. He believed that we should consider
stock prices to be informative if they are between one-half and twice their
fundamental values! Most economists believe that the prices of actively traded
securities are well within these extreme bounds, but no one can know for sure.
• Source: Fischer Black, 1986, “Noise,” Journal of Finance 41(3), 529-543
• Changes in fundamental values are completely unpredictable. Since fundamental
values reflect all available information, they change only when traders learn
unexpected new fundamental information. If fundamental value changes were
predictable, current fundamental values would not fully reflect the information
upon which the predictions are based. Fundamental value changes therefore must
be unpredictable. Since prices are very close to fundamental values in efficient
markets, price changes in efficient markets are quite unpredictable.
• When traders cannot predict future price changes, statisticians say that prices
follow a random walk. Plots of random walks through time look like paths that
wander up or down at random because random walks are completely
unpredictable.
• Informed traders estimate fundamental values. They may base their estimates on private information that only
they have, or on public information that any trader can obtain. Informed traders compare their value estimates
with the corresponding market prices. They consider instruments to be undervalued if prices are less than
their estimates of fundamental value, and overvalued if prices are greater.
• Informed traders buy instruments that they believe are significantly undervalued and sell instruments that
they believe are significantly overvalued. They hope to profit when the prices of their purchases rise, and
when the prices of their sales fall. Informed traders naturally hope that these price changes will occur
quickly.
• Informed traders lose money when they estimate fundamental values poorly. When their value estimates are
wrong, they pay too much for instruments they have overvalued, and they sell toocheaply those instruments
that they have undervalued. Informed traders who consistently estimate values poorly usually quit trading
when they have lost more money than they can tolerate or when bankruptcy forces them out of the markets.
• Informed traders also can lose money even if they accurately estimate
fundamental values. This happens when prices move away from fundamental
values rather than towards them. These losses, however, tend to be short-term
losses. In the long run, prices usually revert toward their fundamental values so
that well-informed traders ultimately profit.
• Even if prices never adjust to their fundamental values, well-informed traders
who have correctly estimated values still can profit from their trades if they are
patient. When they buy an undervalued instrument, they acquire the rights of
ownership for less than their aggregate value. By holding the instrument, they will
eventually receive the benefits of these rights—typically interest, dividends,
royalties, capital repayments, or liquidating distributions—at a lower price than
they could otherwise obtain them. When they sell overvalued instruments, they
can invest the proceeds in instruments with higher expected rates of return.
• Informed traders, like all other traders, often significantly impact prices when they trade. Their
buying tends to push prices up, and their selling tends to push prices down. Since they buy when
price is below their estimates of fundamental value and sell otherwise, the effect of their trading is
to move prices toward their estimates of fundamental value. Their trading therefore causes prices
to reflect their estimates of fundamental value. When informed traders accurately estimate values,
their trading makes prices more informative.
• Informed traders generally differ in their estimates of value. This often happens when they base
their estimates on different data. Informed traders often trade with each other so that the price
impacts of their trading tend to cancel. The net impact of their trading is a market price that
reflects an average of their different value estimates. This price usually is more informative than
are any of the individual value estimates. Markets aggregate data from many sources to produce
prices that typically estimate fundamental values more accurately than can any individual trader.
Volatality
• Volatility is the tendency for prices to unexpectedly change. Prices change in
response to new information about values and in response to the demands of
impatient traders for liquidity.
• Volatility itself changes through time. Sometimes prices are very volatile. Other
times, prices are very stable and hardly change at all. Large price changes
sometimes occur in short time intervals. Regulators and traders refer to episodes
of such price changes as episodic volatility. Episodic volatility concerns many
people because it can be quite scary.
• Volatility, risk, and profit are closely related. Every drop in prices creates losses
for traders who have long positions and profits for traders with short positions.
Likewise, every price rise causes losses for traders with short positions and
profits for traders with long positions. Traders therefore are very interested in
volatility because it can have a significant impact on their wealth. If risk scares
you or profits interest you, you need to know about volatility.
• Volatility especially concerns options traders. Option contract values depend
critically on the volatility of the underlying instrument. Options traders must be
able to measure and predict volatilities to trade profitably. Both skills require that
they understand well the origins of volatility.
• Technical traders who try to interpret trading volumes also pay close attention to
volatility because volumes and volatility are often correlated. The relation
between the two variables is not simple, however. It depends on the origins of
the volatility.
• Volatility greatly concerns regulators. Excessive volatility may indicate that
markets are not functioning well. Since accurate prices are extremely important in
the economy, regulators pay close attention to the markets when prices are highly
volatile. They are especially attentive when markets crash.
• Volatility is of two types. Fundamental volatility is due to unanticipated changes
in instrument values while transitory volatility is due to trading activity by
uninformed traders.
• The distinction is important both for traders and for regulators. Traders must
distinguish between the two volatility types to accurately predict future volatility,
the profitability of dealing strategies, and transaction costs. Regulators must
distinguish between them because they cannot have any lasting effect on
fundamental volatility, but they often can substantially affect transitory volatility.
Depending on the policies regulators adopt, they may decrease or increase
transitory volatility.
• Since economies use prices to allocate resources, it is very important that prices
reflect fundamental values. Values change when the fundamental factors that
determine them change. Prices therefore should change when people learn that
fundamental factors have unexpectedly changed. Such price changes contribute
to fundamental volatility.
• When new information about changes in fundamental values is common
knowledge, prices may change without any trading. For example, suppose that an
unexpected killer frost descends upon Florida overnight. The morning news will
undoubtedly report the event. The next day, orange juice futures contracts will
open at a much higher price than the last price of the previous trading day.
• When only a few people know new information about changes in
fundamental values, prices generally will change on high trading
volumes. The well-informed traders will trade on their information.
The pressures their trades put on prices will cause prices to change
to reflect the new fundamental values.
• Since informed traders generally hurt dealers, and since dealers
generally do not know when they trade with informed traders, dealers
try to infer information about fundamental values from their order
flows.
Fundamental factors
Any factor that determines the value of a trading instrument can cause
the price of that instrument to change. For a commodity, the most
important factors are cash market supply and demand conditions. Other
important factors are interest rates and storage costs. For a bond, the
most important factors are interest rates and the credit quality of the
issuer. For a stock, the most important factors are quality of
management, the values of the company’s resources and technologies,
the supply and demand conditions in its product markets and in its input
markets, and interest rates. For currencies, the important valuation
factors include national inflation rates, macroeconomic policies, and
trade and capital flows. Unexpected changes in any of these factors
generate fundamental volatility in the instrument
Expected changes in fundamental factors generally do not change
prices. Informative prices usually fully incorporate all available
information about future values. Since people base their expectations on
existing information, fully informative prices will already incorporate
expected changes in fundamental factors. When the expected event
occurs, it is not surprising, and it therefore should not cause prices to
change. Only unexpected events cause fundamental price volatility.
Consequently, the identifying characteristic of fundamental volatility in
fully informative prices is unpredictable price changes. An
unpredictable price process is called a random walk.
Bubbles & Crashes
Bubbles and crashes occur when prices greatly differ from
fundamental values. The wealth that these events create,
destroy, and redistribute is often enormous. Bubbles and
crashes thus are quite scary when prices change quickly.
Extreme volatility concerns many people:
 Traders pay close attention to it because large unexpected price changes
expose them to tremendous risks and opportunities.
 Clearinghouses worry about extreme volatility because traders who experience
large losses may be unable to settle their trades or contracts. Clearinghouses
and their members must bear the costs of resulting settlement failures.
• Exchanges and brokers plan for extreme volatility because extreme price changes
usually generate—or are generated by—huge volumes that can overwhelm their
trading systems and cause them to fail. Large sustained price drops especially
concern them because trading volumes usually shrink substantially and remain
low for a long time afterwards
 Microeconomists fret over extreme volatility because very large
price changes often appear inconsistent with rational pricing and
informative prices. They wonder whether excess price volatility
causes people to make poor decisions about the use of economic
resources.
• Macroeconomists fear that the wealth effects associated with large,
broad-based changes in market values may adversely affect the
investment and consumption spending decisions that companies and
individuals make. Poor spending decisions can cause unsustainable
booms and protracted contractions in economic activity
These concerns explain why market regulators regularly
examine trading practices and trading rules that might induce
or attenuate extreme volatility. Some policies that they
consider can create markets that are more resilient. Other
policies have little value, and many policies can harm the
markets. Regulators therefore must carefully analyze how
market structure affects volatility before adopting new policies.
What causes extreme volatility, and how regulations might make it less likely or less
dangerous. Not surprisingly, analysts generally understand the causes of extreme
volatility better after the fact than beforehand. Volatility episodes rarely have
common causes. They do, however, tend to fit a common pattern. Traders who can
recognize conditions that may lead to extreme volatility can take positions that are
highly profitable. Regulators who can recognize these conditions can occasionally
adopt policies to reduce the harmful aspects of extreme volatility.
Bubbles occur when prices rise to levels that are substantially above
fundamental values. (Fundamental values, of course, are not common
knowledge. If they were, crashes and bubbles would not occur.) Some
bubbles occur very quickly. Others occur over long periods. Many
bubbles end with a crash. Traders say that such bubbles pop
Crashes occur when prices fall very quickly. Crashes often follow
bubbles, but they also occur in other circumstances. Crashes sometimes
are called market breaks because the price path breaks when prices fall
very quickly. They also are called market meltdowns when they
overload the order handling capacity of a market.
Bubbles and crashes may affect an individual trading
instrument or many instruments at once. Those that
simultaneously affect many instruments are broad-based
events or market-wide events. Very large price changes most
commonly affect only an individual instrument. Broad- based
bubbles and crashes are quite rare.
•
Bubbles start when buyers become overly optimistic about
fundamental values. The potential of new technologies and the
potential growth of new markets can greatly excite some traders.
Unfortunately, many of these traders cannot recognize when prices
already reflect information about these potentialities. They also may
not adequately appreciate the risks associated with holding the
securities that interest them. If enough of these enthusiastic traders try
to buy at the same time, they may push prices up substantially.
• The resulting price increases may encourage momentum traders to buy
on the hope that past gains will continue. Some momentum traders
may buy because they hope to obtain the profits that their neighbors
and friends have already earned. If enough traders follow them, they
will realize their hopes. The last buyers, however, will lose badly.
• So what?
Algo trading
• Your Algo is fine
• Your momentum math is right

But you Lost!!

Black Swan
• A black swan event is a very rare event
• It happens randomly
• All models are fooled by randomness
• Paranoid people recognize it
• Order anticipators may also buy in anticipation of new uninformed
buyers. They will profit if they can get out before prices fall.
• The combined trading of these traders can cause a bubble in which
prices exceed fundamental values. Momentum traders and order
anticipators, in particular, tend to accelerate price changes. Prices also
accelerate when early buyers grow more confident as their wealth
increases, and when early sellers repurchase their positions to stop
their losses.
• Value traders and arbitrageurs may recognize that prices exceed values, but they may be unable or
unwilling to sell in sufficient volume to prevent the bubble from forming. These traders may be
unable to sell as much as they want to sell if they do not have large positions to sell, if they do not
have enough capital to carry large short positions, or if they cannot easily sell short. They may be
unwilling to sell if they suspect that uninformed traders will continue to push prices up, or if they
lack confidence in their abilities to estimate values well.
• Eventually prices rise to a level that causes sellers to start trading aggressively. The sellers may
be long-term holders, early buyers who want to realize their gains, contrarians, value traders, or
arbitrageurs. Once their selling causes prices to fall, momentum buyers lose their interest.
• Overly optimistic buyers lose their confidence and sellers become more confident. Late buyers
especially worry about their positions and often start selling to stop their losses. Those traders
who financed their positions on margin may have to sell their positions to satisfy margin calls
from their brokers. Other long holders who have placed stop loss orders also will start to sell.
Order anticipators may anticipate these margin calls and these stop orders and sell before them. A
crash occurs when the combined effect of all their selling causes prices to quickly fall.
• A very good book for India
• Answers a lot of questions
• Will refer to it often
How do you identify a bubble?
• High PE which is possible during high growth
• Commodity Stocks are rising
• Economic Growth and Stock price rise are disproportional
• Media Hype
• Sudden Entry of Less Known firms
• Entry of more firms in the sector
Lessons
• Valuations (and fundamentals) matter the most
• Avoid triple digit PE’s
• Mid Cap and Small Cap Stocks(Why?)
• Avoid Large little known companies
• Information Asymmetry- be careful
• This time it is different-be careful
• Never get married to your stocks
• Never short sell
Why do bubbles repeat?
• Demographic Changes
• Short Public Memory
• There is no way of measuring the new Phenomena- Covid??
• Overconfidence of Knowledge, Ability(and tools)- the AI phenomena

Fischer Black on Noise

Fischer Black was a mathematician who made many seminal contributions to the
development of financial theory. Perhaps most notably, he helped develop option-
pricing theory, for which Myron Scholes and Robert Merton received the 1997
Nobel Prize in Economic Science. Had Fischer not died two years before the Prize
was awarded, he undoubtedly also would have been a Nobel laureate.
In his 1985 presidential address to the American Finance Association, Black
offered a now famous opinion about noise. He believed that we should consider
stock prices to be informative if they are between one-half and twice their
fundamental values! Most economists believe that the prices of actively traded
securities are well within these extreme bounds, but no one can know for sure.
• Source: Fischer Black, 1986, “Noise,” Journal of Finance 41(3), 529-543
Capitalization to GDP
• In Jan 2021 India’s market cap crossed 100 % of its GDP
• At the lowest is was 52%
• Median was 76%
• M2B Vs M2M
• Fundamental value 52/1.5 0r 76/2 say 38%?
Products
Products
• Stocks & Shares
• Bonds
• Derivatives-Forward & Futures,Options & Swaps
• Portfolio(Capital Allocation)
• Index funds
Fundamental
Enterprise/Company Performance
• Good book
• A bit in detail
• Good for beginners
Macro factors
• What business is the company in?
• What geography does it operate in
• What is the TAM?
• Who are competitors?
• What is the USP?
• Who is the Management?
• What is the Quality?
Information Sources
• Quarterly results
• Yearly results
• Balance sheet
• Cash flow statements
• Stautory filing
• Conference calls?
Numbers
• Revenue
• Other Income
• Cost of Operations
• Cost of Sales
• Cash flow
• Tax
• Dividend policy?
Exercise
• Take a company of your choice-listed
• Study its public balance sheet, cash flow
• Study competition
• Why would you invest/not invest?
 10-03-2021
Ratio’s-Operational & Investor , Shares as a product
• Gross profit margin, operating profit margin, and net profit
margin are the three main margin analysis measures that are
used to analyze the income statement activities of a firm.
• Each margin individually gives a very different perspective on
the company’s operational efficiency. Comprehensively the
three margins taken together can provide insight into a
firm’s operational strengths and weaknesses (SWOT). Margins
are also useful in making competitor comparisons and
identifying growth and loss trends against past periods.
Gross Profit Margin
• Gross Profit Marginanalyzes the relationship between gross
sales revenue and the direct costs of sales. This
comparison forms the first section of the income statement.
Companies will have varying types of direct costs depending on
their business. Companies that are involved in the production
and manufacturing of goods will use the cost of goods sold
measure while service companies may have a more
generalized notation.
• Overall, the gross profit margin seeks to identify how efficiently
a company is producing its product. The calculation for gross
profit margin is gross profit divided by total revenue. In general,
it is better to have a higher gross profit margin number as it
represents the total gross profit per dollar of revenue.
Operating Profit Margin
• Operating efficiency forms the second section of a company’s
income statement and focuses on indirect costs. Companies have a
wide range of indirect costs which also influence the bottom line.
Some commonly reported indirect costs includes research and
development, marketing campaign expenses, general and
administrative expenses, and depreciation and amortization.
• Operating profit margin examines the effects of these costs.
Operating profit is obtained by subtracting operating expenses from
gross profit. The operating profit margin is then calculated by dividing
the operating profit by total revenue.
• Operating profit shows a company’s ability to manage its indirect
costs. Therefore, this section of the income statement shows how a
company is investing in areas it expects will help to improve its brand
and business growth through several channels. A company may
have a high gross profit margin but a relatively low operating profit
margin if its indirect expenses for things like marketing, or capital
investment allocations are high.
Net Profit Margin
• Net profit margin is the third and final profit margin metric used
in income statement analysis. It is calculated by analyzing the last
section of the income statement and the net earnings of a company
after accounting for all expenses.
• Net profit margin takes into consideration the interest and taxes paid
by a company. Net profit is calculated by subtracting interest and
taxes from operating profit—also known as earnings before interest
and taxes (EBIT). The net profit margin is then calculated by dividing
net profit over total revenue.
• Net profit spotlights a company’s ability to manage its interest
payments and tax payments. Interest payments can take several
varieties. Interest includes the interest a company pays stakeholders
on debt for capital instruments. It also includes any interest earned
from short-term and long-term investments.
• The net profit margin of a company shows how the company is managing all the
expenses associated with the business. On the income statement, expenses are
typically broken out by direct, indirect, and interest and taxes. Companies seek to
manage expenses in each of these three areas individually.
• By analyzing how the gross, operating, and net profit margins compare to each other,
industry analysts can get a clear picture of a company’s operating strengths and
weaknesses.
• Market and business factors may affect each of the three margins differently.
Systematically if direct sales expenses increase across the market, then a company
will have a lower gross profit margin that reflects higher costs of sales.
• Companies may go through different cycles of growth that lead to higher operational,
and interest expenses. A company may be investing more in marketing campaigns or
capital investments that increase operating costs for a period which can decrease
operating profit margin. Companies may also raise capital through debt which can
decrease their net profit margin when interest payments rise.
• Understanding these different variables and their effects on margin analysis can be
important for investors when analyzing the worthiness of corporate investment.
XBRL
• XBRL stands for eXtensible Business Reporting Language. It is a
language for the electronic communication of business and financial data
which is revolutionising the business reporting around the world. It offers
major benefits to all those who have to create, transmit, use or analyse
such business information. Some of the benefits of using XBRL include
cost savings, greater efficiency, improved accuracy and reliability to all
those involved in supplying or using financial data.
• The basic idea behind XBRL is that instead of treating financial information
as a block of text or numeric items, a unique electronically readable tag is
attached to each individual financial term. It is not just the data or text that
floats around, these individual items move along with an electronic tag.
Thus, it is not just the 'content' but also the ‘context’ is being transmitted. It
is one of a family of ‘XML’ languages which is becoming a standard means
of communicating information between businesses and on the Internet.
EBITDA, EIT and net Margin
• EBIT (Earnings Before Interest and Taxes) is a proxy for
core, recurring business profitability, before the impact of capital
structure and taxes.
• EBITDA (Earnings Before Interest, Taxes, Depreciation, and
Amortization) is a proxy for core, recurring business cash flow
from operations, before the impact of capital structure and
taxes.
• And Net Income represents profit after taxes, the impact of
capital structure (interest), AND non-core business activities.
Example
• You have to program your algo to choose among 6 companies
manufacturing a widget X
• You can choose to trade only with 2 of the stocks
Parameter Company1 Company 2 Company3 company4 company5 Company 6

Revenue 1000 1500 1300 1800 1200 1100

COGS 650 1050 650 1170 400 500

OM 350 450 650 680 800 600

SG&A, Dep, 150 250 125 540 400 350

R&D
OM(EBIT) 200 200 525 140 400 250

Interest & Tax 60 60 150 40 150 150

NP 140 140 375 100 250 100

Accounting
• Accounting is the basic building block
• Auditors are classified by reputation-Big4-PwC, D&T,EY, KPMG
• Accounting & Reporting Standards
• IFRS
• Int AAS
• IndAS
• XBRL
Stock market scams
• Enron-Sarbanes Oaxley, M2M, Subsidiary, Jeff Skilling, Arthur
Andersen
• Worldcom
• Satyam
• NSEL-FTL
• A good book
• Gives simple definition of ratio’s
Investing Ratio’s
Profitability Ratios
EPS-Earning Per Share
• EPS=Net Profit/No of shares issued
• Minority Interest-Yahoo & Alibaba 
• Weighted Average
• Diluted EPS
• Retained Profit
• Dividend policy
• Convertible debentures
Significance
• Lag indicator
• Indicates ROI
• Helps calculate P/E ratio
• Watch for changes in EPS
Price to Earnings Ratio(P/E)
• P/E= Share Price/EPS
• EPS already explained
• Share price is the mid-market price price of the share at the close of
business on the previous trading day
• Most common ratio for investors
• Sometimes called PER
• Called Trailing P/E
• P/E TTM
• No negative P/E
• Earning multiples- time taken to earn back that share?
• Only listed companies have a P/E. PrivateEquity firms use it as a
surrogate indicator
What does P/E indicate?
• Stock market confidence(and bubble)..Below 10, 10-18,18-30,
upwards of 30.
• Below 10- lack of confidence
• 10-18-rightly priced
• 18-30-Exceptional confidence
• Upwards of 30- correction may be due? Bubble?
• P/E of various sectors
• P/E used to compare with various instruments
Variations to P/E
• PEG=Price to Earnings to Growth Ratio. Calculated by divididing P/E
by growth rate
• Forward PE or estimated P/E. Uses the next year EPS
• P/E from continued operations
• P/E 10. Replace EPS for one year with Average EPS for 10 years
• E/P –simply reverse. Allows comparison of investment
• Market P/E- on a weighted basis for a market/sector. Used for index
A not very popular alternative
PVGO=Present Value of Growth Opportunities=

{D/(r-g)}-E/r

Where
D=Dividend for the next period
r=cost of capital or capitalization rate of the company
E=EPS
G=growthrate of the company
EV/EBITDA
• Enterprise Value is the current value of the business taking debt into
account
• Enterprise Value=Market Capitalization +Total Debt- Total Cash
• Market Capitalization= Market Price per share x No of shares
• EBITDA can be calculated from he balance sheet
• Debt is added on the EV side and Debt on the EBITDA side
• Some promoters prefer debt, This ratio brings all to comparable basis
• EBIDTA indicates that Net Profit may not be positive
• EV/EBIDTA is for companies which are not start-up’s but not earning
enough for a net profit.It is a modified form of P/E for companies yet
to generate a profit.
• It has to be read with other metrics
• Limited Vue
Price-to-Sales Ratio(PSR)
• Price to Sales Ratio=Market Cap/Annual Sales
• Revenue is same as Annual Sales
• Limited Use
• Used when profits are not a good metric-retail
• Used along with other metrics
The 4 metric discussed so far are Profitability Ratios
• EPS
• P/E
• EV/EBITDA
• PSR
Investment Ratio’s
• Dividend Yield
• Total Return
• Return on Capital Employed
• Return on Equity
• Premium on Asset Value
• Internal Rate of Return
• Return of Assets
• Dividend Payment Ratio
Dividend Yield
• Dividend Yield=Dividend/Share Price
• Tax
• Dividend Record
• Dividend Yield
• May not be applicable
• Dividend Policy
 Total Return
• Total Return=Capital Growth+Dividend yield at purchase price
• Capital Growth is the increase in value of shares measured as a
percentage of the amount paid for them
• Dividend yield is calculated like earlier except that we use purchase
price instead of market price
• Return is a inflated figure for long held shares-inflation
• Return on Current Value-Divide it by number of years or take the root
of the period
ROCE(Return on Capital Employed)
• ROCE=(EBIT/Net Capital Employed) x100
• Net Capital Employed is taken from the balance sheet.Simplest way to
add all assets and subtract current labilities.
• NOPLAT-Net Operating Profit Less Adjusted Taxes
• Use of NOPLAT to get comparison of cost of Capital
• ROCE on trading
Return On Equity
• Return on Equity=(Net profit attributable to shareholders/equity)x100
• Net profit attributable to shareholders is the net profit after deduction of
interest,tax and all other items expect dividends(paid and payable).This
figure is available to shareholders
• Equity in this context means value of the assets which may regarded as
owned by the shareholders.This figure is sometimes called shareholder
funds,equity or net assets
• Shareholders equity=Fixed Assets+current assets-current liabilities-long
term creditors-provisions
• Goodwill
• Varies from industry.15-20% ?
Premium to Asset Value(PAV)
• Premium to asset value={(share price-net asset value)x100}-100
• NAV=(Shareholder funds-goodwill)/number of shares in issue
• If large difference between NAV and PAV- maybe undervalued?
• PAV is a means to check premium
Internal Rate of Return(IRR)
• It is the overall rate of return from more than one variable
• Not a ratio in the rue sense
• =XIRR is thefunction on Mircosoft Excel
• It gives you the rate of return on shares compared to other
instruments
• It is a compounding rate of interest.Presently 8% is good, 12% is very
good.
• Can compare with yields on bonds
Return on Assets
• ROA=(Net Pofits/Net Assets)x100
• Net profit is profit before taxes and dividends
• Net assets is total of all fixed assets(less depreciation) plus current
assets, minus current liabilities
• ROA is a % which represents the return the company is itself
makingon what it owns
• Bankers use this
• Variations
Dividend Payout Ratio
• Dividend Payout ratio=Dividend per share/EPS
• Dividend per share is the total of the year’s dividend payablefor
owning one share.If a company pays an interim and final dividendthey
must be added.The amount included is the amount received by the
shareholder.No adjustment made for tax
• Retained Earning ratio is the opposite
• Useful as an investor for trusts etc
Dividend Cover
• Dividend Cover=EPS/Dividend per share
• It gives out how sustainable the dividend is
• Usually should be 3 times-last 3 years
• Below 1 is a warning
Overheads to Turnover Ratio
• Overheads/Turnover=(overheada/turnover)x100
• Overheads are all the expenses or indirect costspf the company.They
are the the items between GP and NP
• Turnover is Sales
• This has to reduce as company grows
• Expense Ratio is the same
13-03-2021
Ratio’s and Products
 REGIONAL HEAD OF ADVISORY SOLUTIONS AND PRODUCT MANAGEMENT
• EFG BANK • 2012 TO 2020
Member of Singapore Management Committee. As a member of the Asia Business
Committee, help establish strategic directions for the Asian region. In the role of Regional
investment head, oversaw CHF
5. bn in client Assets Under Management with a regional team of Investment
Counsellors, asset class specialists & product management experts. A regular
speaker at various investment forums.

HEAD OF MARKET SPECIALISTS •COUTTS & CO• 2006 TO 2012

Establish the cross functional product specialist team in Coutts providing advisory to high
net worth individual clients in equity, fixed income, forex, third party funds. Represented
Coutts in investment seminars.

DIRECTOR (EQUITIES, ASIA PACIFIC TEAM) •UOB ASSET MANAGEMENT • 2004 TO

2006
As a fund manager, managed a global internet fund and the Taiwan portion of the Asia
Pacific Funds under UOBAM’s ambit. Concurrently senior analyst covering the technology
sector.
TERENCE TAN
ASSOCIATE DIRECTOR, HEAD OF REGIONAL TECHNOLOGY RESEARCH • DBS.

ASSISTANT MANAGER (RESEARCH ANALYST) • YAMAICHI RESEARCH INSTITUTE •

7th and 10th April 1996 TO 1997
Equity analyst covering Asia region in the food & beverage and shipping sectors.

COUNTRY OFFICER (INDONESIA) • MINISTRY OF FOREIGN AFFAIRS • 1995 TO

1996
CAPTAIN (NS), SECTION HEAD •HQ RSAF •1983 TO 1995
More Ratio’s
Gearing Ratio
• Gearing ratio=(Total borrowings-cash)/shareholder funds.Can also be expressed
as a %
• Ungeared,Low-Geared, High-Geared, Medium-Geared
• It has a magnifying effect on EPS
• High Gearing has a dramatic effect on EPS, hence the term gearing
• High gearing can turn a profit into a loss for the ordinary shareholder especially
when profits are muted
• Different industries-airline,hotel,power-different gearing
• Types of Debt equity
• Risk appetite
• Organic growth
• Which company dramatically lowered their gearing in the pandemic?
Interest Cover(age)
• Interesr Cover=(Pretax fofit-interest paid)/interest
• Gives an idea of how many times interest is earned as profit
• Minimum should be >3
• Interest sensivity in indicated by interest cover. If interest rates
increase profitability should not be grossly affected
• Bonds-Interest Covenants
• Notes
Solvency Ratio’s
Acid test
• Bad Bank
• ARC
• Trading opportunity
• Acid test ratio=(current assets-stock)/current liabilities
• Nitric Acid-Gold rush
• Also called liquidity ratio or quick ratio
• Also influenced by cash flow
• Retail, restraunts are cash positive, real estate cash negative
Current Ratio
• Current Ratio=Current Assets/Current Liabilities
• Use the Acid ratio or Current ratio, not both
• If stock can be quickly converted to Assets use Current Ratio
• “Profit is an opinion, cash is reality”
Cash Burn
• Also called Burn Rate
• Cash Burn= (Capital at the start of the year)/Year’s cash expenditure
• Capital at the start of the year is the amount of cash(not debtors not
stock) at the beginning of the financial year. It is read from the Cash
flow statement
• Years cash expenditure is the amount of cash spent during the year
excluding cash from profits and interest received. Again CF statement
• Used for new companies, ARC, refinanced companies
Defensive Interval
• Defensive Interval=Liquid Assets/Daily operating expenses
• Liquid Assets are cash, debtors and stock if readily saleable
• Daily operating expensesare the cost of running the business for one
day.It is every aspect and cost of the business. Mathematically is is
everything that was subtracted from Sales for Net Profit before Tax.It
includes interest payments
• It gives how much a business can sustain without income-30-90 days
Fixed Charges Cover
• Fixed Charges Cover=Net Profit before fixed charges/fixed charges
• Fixed charges is the charges that the organization must pay
irrespective of whether it is functioning-rent,Property tax, lease etc
• Indicates how they can play out a recession
Efficiency Ratio’s
Stock Turn
• Sales/Closing Stock
• Not applicable in non manufacturing industries
• Different industries, different ratio’s
• Stock Days=(Stock/Sales)x365
Price to Book Value
• M2M and M2B
• PBV=Market Capitalization/Shareholder’s equity
• Market capitalization is the current market value of all the shares
• Shareholder assets is the value of the business as shown by the net
tangible assets or net assets
• Has limited value
Overtrading and Undertrading
• Overtrading=Capital required/Capital Available
• Capital required is a subjective measure of how much capital is
required to operate
• Capital required is all the capital available-shares, loans etc
• Overtrading is that company has less funds than that required
• Other indicators are available
• Undertrading is not some much of a concern
Item Comparison
• Item Comparision=Item in accounts for current year/same item for
previous year
• Intercompany Comparison
• Divisional Comparison
• Used by investors to find sudden changes
• Make Comparisions
Policy Ratios
Creditor Period
• Creditor Period=(Trade Creditors/Cost of Sales) x365
• Only those Creditors company has a discretion on are included
• Gives investors a sense of use of working capital
Debtor Period
• Debtor Periods=(Trade Debtors/Sales)x365
• Known as DSO
• Very important for engineering and other industries
• Impacts WC
Fixed Asset Spending Ratio
• Fixed Asset Spending Ratio=Fixed Asset Spending/annual depreciation
• Fixed asset spending is the amount company has spent on Fixed
Assets during the last one year and are expected to last more than a
year
• Depreciation- useful life-straight line
• Land-never depreciated
• Revaluation
Volatility
Volatility Ratio
• Volatility Ratio={(Period high-Period Low/2)/current share price}x100
• Period high & Period Low Prices are the high and low prices
• High volatility is a cause of concern
Standard Deviation

2
𝑥 − 𝑥ҧ
𝜎=
𝑛
• 𝜎=standard deviation
• 𝑥=the average arithmetic mean share price
• 𝑥=the
ҧ share price at the end of the day
• n=number of share prices ,sometimes denoted as n-1
• 1.Calculate the amount by which each day’s closing price differs from
the average
• 2.Square each answer
• 3.Add up the squares
• 4.Divide by the number of answers
• 5.Take the square root
• Share A 96,101,103 and 100.Drawing a bell curve and calculating SD
SD IS 2-45.There is a 68.2% probability that A will differ by 2.45
Share B -80.92.130.98-COMMENT?
• SD=18.49
• 2X18.49=36.98
• Probability=95.4%
• STDEVon excel
• Mean Varaince optimization
• Risk Reward
 What did we cover
• Profitability Ratio’s • Margins • .Efficiency Ratio
• EPS • GM,OM,Net • Stock Turn
• P/E • OH on Turnover • Pbv
• EV • Gearing • Over trading and
• ESR • Gearing Ratio Under Trading
• Investment Ratio’s • Interest Cover • Policy Ratio
• Dividend Yield • Solvency Ratio • Debtor Period
• Total Return • Acid test
• ROCE • Creditor Period
• Current Ratio • Fixed Asset
• ROE • Cash Burn
• PAV • Defensive Interval Spending Ratio
• IRR • Fixed Charges Cover • Volatility
• ROA • Volatility Ratio
• Dividend Payout Ratio • Standard Deviation
• Dividend Cover
In Summary
• Enterprise/Company is the bedrock of trade
• P&L and Cash flow are the mathematical measures from accounting
• Ratio’s are derived from these
• Each investor makes his own thesis based on ratio’s
In God we trust-the rest bring numbers
Instruments
Stocks
Growth of a company/Enterprise
• Propietership
• Partnership
• LLP
• Pvt Ltd
• Closely Held Publicly Limited
• Listed
What are Shares?
• Shares are units of equity ownership interest in a corporation
that exist as a financial asset providing for an equal distribution
in any residual profits, if any are declared, in the form
of dividends. Shareholders may also enjoy capital gains if the
value of the company rises.
• Shares represent equity stock in a firm, with the two main types
of shares being common shares and preferred shares. As a
result, "shares" and “stock" are commonly used
interchangeably.
• In India they are called promoter’s shares
Key Takeaways
• Shares represent equity ownership in a corporation or financial
asset, owned by investors who exchange capital in return for
these units.
• Common shares enable voting rights and possible returns
through price appreciation and dividends.
• Preferred shares do not offer price appreciation but can be
redeemed at an attractive price and offer regular dividends.
• Most companies have shares, but only the shares of publicly-
traded companies are found on stock exchanges.
Unlike debt capital, obtained through a loan or bond issue, equity
has no legal mandate to be repaid to investors, and shares, while
they may pay dividends as a distribution of profits, do not pay
interest. Nearly all companies, from small partnerships or LLCs to
multinational corporations, issue shares of some kind.
• Shares of privately held companies or partnerships are owned
by the founders or partners. As small companies grow, shares
are sold to outside investors in the primary market. These may
include friends or family, and then angel or venture (VC)
investors. If the company continues to grow, it may seek to raise
additional equity capital by selling shares to the public on
the secondary market via an Initial Share Offering (IPO). After
an IPO, a company's shares are said to be publicly tradedand
become listed on a stock exchange
Authorized and Issued Shares
• Authorized shares comprise the number of shares a company’s
board of directors may issue. Issued shares comprise the
number of shares that are given to shareholders and counted
for purposes of ownership.
• In India SEBI controls the issue of shares
• Bonus Shares-Dilutes Value
• Stock Split-Does not
• For some BOD is sufficient
• For some a AGM
Rights
• Most companies issue common shares. These provide
shareholders with a residual claim on the company and its
profits, providing potential investment growth through both
capital gains and dividends. Common shares also come
with voting rights, giving shareholders more control over the
business. These rights allow shareholders of record in a
company to vote on certain corporate actions, elect members to
the board of directors, and approve issuing new securities or
payment of dividends. In addition, certain common stock comes
with pre-emptive rights ensuring that shareholders may buy new
shares and retain their percentage of ownership when the
corporation issues new stock.
In comparison, preferred shares typically do not offer much
market appreciation in value or voting rights in the corporation.
However, this type of stock typically has set payment criteria, a
dividend that is paid out regularly, making the stock less risky
than common stock. Because preferred stock takes priority over
common stock if the business files for bankruptcy and is forced to
repay its lenders, preferred shareholders receive payment before
common shareholders but after bondholders. Because preferred
shareholders have priority in repayment upon bankruptcy, they
are less risky than common shares.
Bonds
What is a Bond?
A bond is a fixed income instrumentthat represents a loan made
by an investor to a borrower (typically corporate or
governmental). A bond could be thought of as an IOU between
the Lender and borrower that includes the details of the loan and
its payments. Bonds are used by companies, municipalities,
states, and sovereign governments to finance projects and
operations. Owners of bonds are debtholders, or creditors, of the
issuer. Bond details include the end date when the principalof the
loan is due to be paid to the bond owner and usually includes the
terms for variable or fixed interest payments made by the
borrower.
Key takeaway’s
• Bonds are units of corporate debt issued by companies and
securitized as tradeable assets.
• A bond is referred to as a fixed income instrument since bonds
traditionally paid a fixed interest rate (coupon) to debtholders.
Variable or floating interest rates are also now quite common.
• Bond prices are inversely correlated with interest rates: when
rates go up, bond prices fall and vice-versa.
• Bonds have maturity dates at which point the principal amount
must be paid back in full or risk default.
Who issues them?
Governments (at all levels) and corporations commonly use
bonds in order to borrow money. Governments need to fund
roads, schools, dams or other infrastructure. The sudden
expense of war may also demand the need to raise funds.
Funding wars…
The Rothschild family are known to have funded the losers and the
winners of all wars since the Napoleonic Wars of 1803 to 1815. They
fund both sides so that whichever side wins, they profit. In 1815,
Napoleon Bonaparte was campaigning to take over Britain itself.
Seeing an opportunity, the Rothschild family got together and made a
plan. Jacob Rothschild was stationed in Paris, France. He was told to
fund Napoleon’s war effort to conquer Britain. On the other hand,
Nathan Rothschild, stationed in London was told to fund General
Wellington’s war effort against Napoleon. The two sons did that and
the war raged. Soon, the family decided that it was more profitable to
them if Britain won the war and the British royal family was indebted to
the Rothschild family. Thus, Jacob limited the funds to Napoleon’s
army and General Wellington began to win.
India
• Jagat Seth
• Battle of Plassey
• Robert Clive
Similarly, corporations will often borrow to grow their business, to
buy property and equipment, to undertake profitable projects, for
research and development or to hire employees. The problem
that large organizations run into is that they typically need far
more money than the average bank can provide. Bonds provide
a solution by allowing many individual investors to assume the
role of the lender. Indeed, public debt markets let thousands of
investors each lend a portion of the capital needed. Moreover,
markets allow lenders to sell their bonds to other investors or
to buy bondsfrom other individuals—long after the original issuing
organization raised capital
• Bonds are commonly referred to as fixed income securities and
are one of three fundamental asset classes individual investors
are usually familiar with, along with stocks (equities) and cash
equivalents.
• Many corporate and government bondsare publicly traded;
others are traded only over-the-counter (OTC) or privately
between the borrower and lender.
•
When companies or other entities need to raise money to finance
new projects, maintain ongoing operations, or refinance existing
debts, they may issue bonds directly to investors. The borrower
(issuer) issues a bond that includes the terms of the loan, interest
payments that will be made, and the time at which the loaned
funds (bond principal) must be paid back (maturity date). The
interest payment (the coupon) is part of the return that
bondholders earn for loaning their funds to the issuer. The
interest rate that determines the payment is called the coupon
rate.
The initial price of most bonds is typically set at par, usually $100
or $1,000 face value per individual bond. The actual market price
of a bond depends on a number of factors: the credit quality of
the issuer, the length of time until expiration, and the coupon rate
compared to the general interest rate environment at the time.
The face value of the bond is what will be paid back to the
borrower once the bond matures.
Most bonds can be sold by the initial bondholder to other
investors after they have been issued. In other words, a bond
investor does not have to hold a bond all the way to its maturity
date. It is also common for bonds to be repurchased by the
borrower if interest rates decline, or if the borrower’s credit has
improved, and it can reissue new bonds at a lower cost.
Advantages and Disadvantages
• Income through interest but lower than stock, taxes
• Get your principal back, but companies can default
• You can sell the bond at higher price but yield may fall
Some terms
• Face value is the money amount the bond will be worth at
maturity; it is also the reference amount the bond issuer uses
when calculating interest payments. For example, say an
investor purchases a bond at a premium $1,090 and another
investor buys the same bond later when it is trading at a
discount for $980. When the bond matures, both investors will
receive the $1,000 face value of the bond.
• The coupon rate is the rate of interest the bond issuer will pay
on the face value of the bond, expressed as a percentage. For
example, a 5% coupon rate means that bondholders will receive
5% x $1000 face value = $50 every year
Coupon dates are the dates on which the bond issuer will make
interest payments. Payments can be made in any interval, but
the standard is semiannual payments.
The maturity date is the date on which the bond will mature and
the bond issuer will pay the bondholder the face value of the
bond.
The issue price is the price at which the bond issuer originally
sells the bonds.
What affects a bond price?
• Credit Quality and Duration
• Credit rating agencies
• Investment grade
• Junk Grade
• Bonds and bond portfolios will rise or fall in value as interest
rates change. The sensitivity to changes in the interest rate
environment is called “duration.” The use of the term duration in
this context can be confusing to new bond investors because it
does not refer to the length of time the bond has before
maturity. Instead, duration describes how much a bond’s price
will rise or fall with a change in interest rates
• The rate of change of a bond’s or bond portfolio’s sensitivity to
interest rates (duration) is called “Convexity”.
• Convexity of a Bond is a measure that shows the
relationship between bond price and Bond yield, i.e., the
change in the duration of the bond due to a change in the
rate of interest, which helps a risk management tool to
measure and manage the portfolio’s exposure to interest
rate risk and risk of loss of expectation
Bonds & Derivatives

Class 7-15-03-2021
Time Value of Money(TVM)
• The Time Value of Money (“TVM”) is a concept on which the
rest of finance theory rests on. Therefore, it is critical to
understand this concept well.
• Rs 100 today will buy a lot less ten years from now and brought
a lot more 10 years back
• As time flows value of money declines
• The value of money declines due to the combined impact of the
following:
1.Inflation in the economy;
2.Risks involved in delayed receipts of cash or financial
transactions; and
3.Opportunity cost of capital delayed.
• While each of these forces alone can cause the value of money to
decline individually, all the three usually act with different
degrees of impact to cause a decline in the value of money as
time flows.
 The Present Value Formula
• The present value formula quantifies how fast the value of money declines. This formula
shows you how much once single cash payment (FV) received in a future time period (t) is
worth in today’s terms (PV).

PV= FVt
__________________

(1+r)t

• Present Value (PV) stands for the value of the money in today’s terms.

• Future Value (FV) stands for the amount of cash received in the future.
r is the discount rate or the speed at which the decline in value is happening
• t is the time period in which the future value or cash is received.
Future Value of a Sum
The future value of a sum depends on the interest rate and the
interval of time over which the sum is invested. This is shown
with the following formula:
FVt = PV*(1+r)t
• where:
• FVt = future value of a sum invested for t periods
• r = periodic interest rate
• PV = present value
• t = number of periods until the sum is received
Example
• A sum of Rs 1000 is invested at a rate of interest of 3% for 4 years.
What is the value of the sum at the end of 4 years?
• PV=1000, t=4 and r=3
• FV4 = 1,000(1+.03)4
• FV4 = 1,000(1.12551)
• =1125.51 Rs
Example 2
• How much must be deposited in a bank account that pays 5%
interest per year in order to be worth Rs1000 in three years?
• In this case, t = 3, r = 5% and FV3 =Rs 1000.
• PV=1,000/(1.1576)
• =Rs 863.84
Interest Rate and Compounding Frequencies
• Simple
• Annual Compounding
• Semi-Annual Compounding
• Monthly Componding
• t changes in value
• Take 1000 Rs invested for 2 years at 8%
• Annual=1,000*(1+.08)^2=1166.4
• Semi Annual=1,000*(1+.04)^4=1169.86
• Monthly=1,000*(1+.006667)^24=1172.89
Annuities
• An annuity is a periodic stream of equally-sized payments. The
word annuity is derived from the Latin word annum (yearly). In
spite of this, any stream of periodic payments of equal size can
be treated as an annuity. As an example, mortgage payments are
made monthly and are of equal size, and so can be thought of as
a type of annuity.
The two basic types of annuities are:
• Ordinary annuity
• Annuity due
Ordinary Annuinity
With an ordinary annuity, the first payment takes place one
period in the future. Most annuities are ordinary; some examples
are:
• Coupons paid by a bond
• Dividend payments by a share of preferred stock
• Car loan payments
• Mortgage payments
• Student loan payments
• Social security payments
The Future
Value of an FVAt=Future value of t-
Ordinary period of an annuity

Annuity C=Periodic cash flow

r=periodic interest rate

t=number of periods until

the sum is recieved
• As an example, suppose that a sum of Rs1,000 is invested each
year for four years, starting next year, at an annual rate of
interest of 3%. Since the cash flows start next year, this is an
ordinary annuity. What is its future value? In this case, t = 4, r =
3% and C = Rs1,000.
• Alternatively, the future value of each individual cash flow can
be computed and then combined as follows
• The first cash flow is invested for three years (from year one to
year four):
• FV3 = PV(1+r)t
FV3 = 1,000(1+.03)3
FV3 = 1,000(1.09273)
FV3 = Rs1,092.73
• The second cash flow is invested for two years (from year two to
year four):
• FV2 = PV(1+r)t
FV2 = 1,000(1+.03)2
FV2 = 1,000(1.06090)
FV2 = Rs1,060.90
• The third cash flow is invested for one year (from year three to
year four):
• FV1 = PV(1+r)t
FV1 = 1,000(1+.03)1
FV1 = 1,000(1.03)
FV1 = Rs1,030.00
• The fourth and final cash flow does not earn any interest since it
is not deposited into the bank until year four. The future value is
therefore Rs1,000
The sum of these future values is:
1,092.73 + 1,060.90 + 1,030.00 + 1,000.00 = Rs4,183.63
The Present The formula for computing
the present value of an
Value of an ordinary annuity is:
Ordinary PVAt = present value of a t-
period ordinary annuity
Annuity C = the value of the periodic
cash flow
r = periodic interest rate
t = number of periods until
the sum is received
• As an example, how much must be invested today in a bank
account that pays 5% interest per year in order to generate a
stream of payments of Rs1,000 in each of the following three
years? In this case, t = 3, r = 5% and C = Rs1,000.
Alternatively, the present value of each individual cash flow can
be computed and then combined as follows:
• The present value of the first cash flow (paid in one year) is:
• PV = FVt / (1+r)t
PV = 1,000 / (1+.05)1
PV = 1,000 / 1.05
PV = Rs952.38
• The present value of the second cash flow (paid in two years) is:
• PV = FVt / (1+r)t
PV = 1,000 / (1+.05)2
PV = 1,000 / 1.10250
PV = Rs907.03
• The present value of the third cash flow (paid in three years) is:
• PV = FVt / (1+r)t
PV = 1,000 / (1+.05)3
PV = 1,000 / 1.15763
PV = Rs863.84
The sum of these present values is:
Rs952.38 + Rs907.03 + Rs863.84 = Rs2,723.25
Annuities Due

With an annuity due, the first payment takes place immediately.

This is a less common type of annuity than the ordinary annuity.
An example of this would be a lease agreement or a loan where
the first payment is due immediately.
Due to the timing of the cash flows, the present value and future
value of an annuity will be affected by whether the annuity is an
ordinary annuity or an annuity due.
Lehman Bros
• Mortgage Loans
• Biggest Bankruptcy
• Derivatives
Term Structures, Interest Rates & Yield Curves
The terms “Term Structure of Interest Rates” and “Yield Curves”
intimidates most people. The concepts of term structure of
interest rates and yield curves intimidates most of us is because
because almost all of us encounter it in their finance courses but
do not go deep into understanding what the term structure or
yield curve ares, how interest rates, yield curves and term
structure are related, what drives interest rates, how the term
structure or yield curves is built and interpreted and most
importantly how the insights gained from the term structure is
used.
Introduction: Term Structures, Interest Rates
and Yield Curves
The term structure of interest rates refers to the relationship
between the yields and maturities of a set of bonds with the same
credit rating. Typically, the term structure refers to Treasury
securities but it can also refer to riskier securities, such as AA
bonds. A graph of the term structure of interest rates is known as
a yield curve.
For example, the following table shows the
term structure of interest rates for Treasury
securities -2016 US
MATURITY YIELD

1 month 0.25%

3 months 0.26%

6 months 0.35%

1 year 0.45%
2 years 0.61%
3 years 0.71%
5 years 1.00%
7 years 1.26%
10 years 1.46%
20 years 1.83%
30 years 2.27%
The corresponding yield curve
Historically, the yield curve has been upward-sloping. This is the
how the yield curve normally looks, it has been referred to as the
‘normal yield curve’. The yields of longer-maturity bonds tend to
be higher than the yields of shorter-maturity bonds since the
longer maturity bonds are riskier. This is because
• Changes in market conditions have a greater impact on the
prices of longer maturity bonds than shorter maturity bonds;
and
• There is more uncertainty over market conditions that take
place further in the future.
The shape and level of the yield curve can change over time. If
economic activity is expected to accelerate in the future, the yield
curve tends to become steeper, since future rates are expected to
be higher than they are now. If the economy is expected to slow
in the future, the yield curve tends to become flatter, since future
rates are expected to be lower than they are now.
During periods of rising inflation, the entire yield curve
shifts up as lenders require higher rates of return to compensate
for the loss of purchasing power. During periods of falling
inflation, the entire yield curve shifts down.
Occasionally, the yield curve can have a negative slope (referred
to as an ‘inverted yield curve’); this tends to happen when
inflation is at unusually high levels and is expected to fall in the
future or when recessions are imminent
Definition of Interest Rates
Interest rates can be expressed in several different equivalent ways,
such as:
• Discount factors
• Spot rates
• Forward rates
• Yields
The prices of Treasury securities may be used to compute discount
factors, spot rates, forward rates and yields. Discount factors can be
computed directly from the prices of Treasuries. Spot rates can be
computed from discount factors; forward rates can be computed from
spot rates.
Discount Factors

• A discount factor represents the present value of a sum. For

example, if the one-year discount factor is 0.942900, this
indicates that the present value of $ 1 received in one year is $
0.942900. Discount factors cannot exceed 1 and will fall
continuously as their maturity increases.
Discount factors may be computed from the prices and coupons
of Treasury securities. As an example, the following table shows
the prices and coupon rates of off-the-run Treasury securities
with maturities of 6 months, 12 months, 18 months and 24
months. Off-the-run securities are those that have been issued in
the past; the most recently issued securities are said to be on-the-
run. For example, if a one-year Treasury bill was issued six
months ago, it is now considered to be an off-the-run six month
Treasury bill. A newly-issued six-month Treasury bill is
considered to be on-the-run.
Data from US Maturity Coupon Rate Price
treasury
6 months 0.80% $990.00
• Each bond has a face value 12 months 1.40% $975.00
of $1,000 and makes semi-
annual coupon payments
18 months 2.20% $960.00
24 months 2.80% $935.00
When the six month bond matures, it will pay one final semi-
annual coupon along with the bond’s principal or face value of
$1,000. The annual coupon rate is 0.8%, so the annual coupon
payment is (0.008)($1,000) = $8. Each semi-annual coupon
payment will therefore be $8/2 = $4. As a result, the bond will
provide a cash flow of $1,004 in six months.
Since the price of a bond equals the present value of its promised
future cash flows, the following relationship holds for the six-
month bond:
990 = 1004d(0.5)
where:
d(t) = the discount factor with a maturity of t years
According to this equation, the price of the bond ($990) equals
the present value of the remaining cash flow of $1,004. Since this
cash flow occurs in six months, its present value is obtained by
multiplying it by the six month discount factor:
1004d(0.5)
Solving for d(0.5) gives:
990 = 1004d(0.5)
d(0.5) = 990 / 1004 = 0.986056
For the 12 month bond, the following relationship holds:
975 = 7d(0.5) + 1007d(1)
This is because the bond’s price is $975. With a 1.4% annual
coupon, the bond pays two semi-annual coupons each year of
(1.4%/2)($1,000) = $7. This bond will make a coupon payment
in six months, and then a final coupon payment and repayment of
principal when the bond matures in twelve months.
Since d(0.5) = 0.986056,
975 = 7d(0.5) + 1007d(1)
975 = 7(0.986056) + 1007d(1)
975 = 6.9024 + 1007d(1)
968.0976 = 1007d(1)
d(1) = 968.0976/1007
= 0.961368
Using this same approach:
• d(1.5) = 0.928366
• d(2) = 0.882386
These results can be confirmed by pricing the 24 month bond using
these discount factors, as follows. The 24 month bond has four
remaining cash flows. The bond’s annual coupon rate is 2.8%, so that
its semi-annual coupon payments equal (2.8%/2)($1,000) =
$14. Therefore, the price of the bond equals:
P = 14d(0.5) + 14d(1) + 14d(1.5) + 1,014d(2)
P = 14(0.986056) + 14(0.961368) + 14(0.928366) + 1,014(0.882386)
P = 13.8048 + 13.4592 + 12.9971 + 894.7394
P = $935.00005
Except for a small rounding error, this matches the bond’s market
price of $935.00.
Spot Rates

A spot rate of interest is the yield to

maturity of a zero-coupon bond. Spot
rates may be derived directly from
discount factors using the following
formula:

where:
S(t) = t-year spot rate
d(t) = t-year discount factor
t = maturity of spot rate (measured in
years)
Forward Rates

As an example, suppose that a bank currently pays a 5% rate of

interest for one-year time deposits, and 6% for two-year time
deposits. An investor deposits $1,000 in a two-year time deposit;
in two years, he will have a balance of
1,000(1.06)2 = $1,123.60.
• Suppose instead that the investor deposits $1,000 in a one-year
time deposit; at the end of that time, the funds are reinvested in
another one-year time deposit. At the end of two years, the
investor will have:
1,000(1.05)(1 + X)
• The amount that the investor will have in two years cannot be
determined since the one-year rate that will apply in one year is
unknown. The rate of interest that equalizes the return under
both scenarios is known as the one-year rate one year
forward. In other words, this is the one year rate that applies
one year in the future
• The one year rate one year forward can be determined by
equalizing the returns to investing at the two year spot rate for
two years and investing at the one year spot rate for one year
and then reinvesting the proceeds for another year at the one
year rate one year forward:
1,000(1.06)2 = 1,000(1.05)(1+X)
(1.06)2/(1.05) = 1 + X
0.0701 = X
The forward rate of interest can be written as f(t,T), where t is the
time at which the rate begins and T is the time at which the rate
matures. In this example, the one-year rate one-year forward is
written as f(1,2). The one-year rate two years forward is written as
f(2,3); the two-year rate one-year forward is written as f(1,3).
• Forward rates can be computed directly from spot rates. Based
on the previous example, the forward rates can be computed
directly from the spot rates:
S(0.5) = 0.028282
S(1) = 0.039789
S(1.5) = 0.050172
S(2) = 0.063552
For example, the six month forward rate six months forward,
designated f(0.5, 1) is the rate at which the following investments
would provide the same rate of return:
Investing for twelve months at the twelve month spot rate (S(1))
Investing for six months at the six-month spot rate (S(0.5)) and
then reinvesting the proceeds at the six-month forward rate six
months forward (f(0.5, 1))
f(0.5,1) = 2[(1.040185/1.014100) – 1]
f(0.5, 1) = 0.051444
Using this same technique, the six-month forward rate one year
forward is computed as:
•
f(1.5,2) = 2[(1.133292/1.077162) – 1]
f(1.5,2) = 0.104218
These results can be confirmed by pricing the 24 month bond
using these spot rates, as follows. When pricing bonds with
forward rates, each cash flow is discounted by a sequence
consisting of the short-term spot rate followed by the appropriate
forward rates.
•
Yields
• The yield to maturity (YTM) is the rate of interest at which the
market price of a bond equals the present value of its expected
future cash flows.
• Yields cannot be determined algebraically; they must be
computed with a specialized financial calculator or Microsoft
Excel.
As an example, the yields for the four bonds in the previous
example can be computed with Excel’s YIELD function, as
follows:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
• settlement = date on which the bond is purchased (for example,
“1/1/15” represents January 1, 2015)
• maturity = date on which the bond matures
• rate = annual coupon rate
• pr = price (as a percentage of face value); for example, if a bond
has a face value of $1,000 and the price is $950, pr would be set
equal to 95 (i.e., 95% of face value)
• redemption = par or face value (based on 100)
• frequency = coupon frequency
• basis = day count convention
• basis = the day-count convention; the options are:
• basis = the day-count convention; the options are:
• 0 = 30/360 (this is the default or assumed value)
• 1 = actual/actual
• 2 = actual/360
• 3 = actual/365
• 4 = European 30/360
For the six month, 0.80% coupon bond, the yield is computed as:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
= YIELD(“1/1/16”, “7/1/16”, 0.8%, 99, 100, 2, 0)
= 0.028283
= 2.8283%
Note that the settlement and maturity dates must be entered with
quotes. Excel assigns a numerical value to each date based on a
starting value of 1 = January 1, 1900. This value is increased by
one for each day; for example, 2 = January 2, 1900. The
numerical value of 1/1/16 is 42370 and the numerical value of
7/1/16 is 42552. These numerical values may be used for
settlement and maturity.
For the twelve month, 1.40% coupon bond, the yield is computed
as:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
= YIELD(“1/1/16”, “1/1/17”, 1.4%, 97.5, 100, 2, 0)
= 0.039748
= 3.9748%
For the eighteen month, 2.20% coupon bond, the yield is
computed as:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
= YIELD(“1/1/16”, “7/1/17”, 2.2%, 96, 100, 2, 0)
= 0.050011
= 5.0011%
For the twenty-four month, 2.80% coupon bond, the yield is
computed as:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
= YIELD(“1/1/16”, “1/1/18”, 2.8%, 93.5, 100, 2, 0)
= 0.063103
= 6.3103%
The Relationship Between Yields, Spot Rates
and Forward Rates
In this case, the yields are increasing with maturity; i.e., the yield
curve is upward sloping. When this happens:
• The shortest maturity yield matches the shortest maturity spot
rate and forward rate;
• As maturities increase, the spot curve rises above the yield
curve, while the forward curve rises above the spot curve
If the yield curve is downward sloping, the reverse is true:
• The shortest maturity yield matches the shortest maturity spot
rate and forward rate
• As maturities increase, the spot curve falls below the yield curve,
while the forward curve falls below the spot curve
If the yield curve is flat (i.e., the yield is the same for all
maturities), then:
• The shortest maturity yield matches the shortest maturity spot
rate and forward rate
• As maturities increase, the yield curve, spot curve and forward
curve coincide with each other and are flat
Theories of the Term Structure of Interest
Rates
Several theories have been proposed to explain the relationship
between the maturities of bonds and the rates paid by these
bonds. These include:
• Expectations theory
• Liquidity premium theory
• Market segmentation theory
Expectations Theory

• According to the Expectations Theory, long-term rates are an

average of investors’ expected future short term rates of interest.
• According to this theory, if the yield curve is upward sloping,
this indicates that investors expect short-term rates to be higher
in the future. If the yield curve is downward sloping, this
indicates that investors expect short-term rates to be lower in
the future. If the yield curve is flat, this indicates that investors
expect short-term rates to be unchanged in the future.
One of the key assumptions of the Expectations Theory is that
investors do not have any preferences for bond maturities; they
are indifferent between bonds of all maturities. For example, an
investor who needs to save for five years would be indifferent
between:
• Buying a five-year bond
• Buying a ten-year bond and selling it after five years
• Buying a thirty-year bond and selling it after five years
The following numerical example illustrates this idea. Suppose
that the current one-year rate is 9% and the current two-year rate
is 10%. According to the Expectations Theory, the fact that the
two-year rate is higher than the one-year rate implies that
investors expect the one-year rate to be higher next year.
• This can be shown by computing the return to the following
strategies:
• Buying a two year bond
• Buying a one-year bond and then reinvesting the proceeds in
another one-year bond
If the investor buys the two-year bond, his return over the next two
years will be:
(1 + 0.10)2 – 1 = 0.21 = 21%
In order for the investor to be indifferent between the two strategies,
he must expect to earn the same return from both. In order for him to
earn 21% by buying a one-year bond with a 9% rate and then
reinvesting in another one-year bond, the expected rate paid by the
second one-year bond must be:
(1 + 0.09)(1 + X) = (1.10)2
(1 + X) = (1.10)2 / (1.09)
X = (1.10)2 / (1.09) – 1
X = 0.11 = 11%
This shows that if one year rates are 9% and two-year rates are
10%, then according to the Expectations Theory, investors expect
the one-year rate to be 11% next year. When long-term rates
exceed short-term rates, this indicates that investors expect
future short-term rates to be greater than they are
today. Equivalently, if long-term rates are below short-term
rates, according to the Expectations Theory, investors expect
short-term rates to be lower in the future than they are today.
Liquidity Premium Theory

According to the Liquidity Premium Theory, a long-term rate of

interest is an average of short-term rates plus a liquidity
premium. In other words, investors expect to be compensated for
holding long-term bonds instead of short-term bonds as long-
term bonds are perceived to be riskier. This causes the yield
curve to be steeper than it would be under the Expectations
Theory.
• As an example, suppose that the one-year rates over the next
five years are expected to be 5%, 6%, 7%, 8% and 9%,
respectively. The liquidity premium for holding bonds of these
maturities equals 0% for a one-year bond, 0.25% for a two-year
bond, 0.5% for a three-year bond, 0.75% for a four-year bond
and 1.00% for a five-year bond. What are the implied two-year
and five-year rates under the:
• Expectations Theory
• Liquidity Premium Theory
Under the Expectations Theory, the implied two-year rate is
computed as:
(1 + X)2 = (1 + 0.05)(1 + 0.06)
(1 + X)2 = 1.113
1 + X = 1.1131/2
X = 1.1131/2 – 1
X = 0.055 = 5.5%
Under the Liquidity Premium Theory, the implied two-year rate
is computed as:
(1 + X)2 = (1 + 0.05 + 0)(1 + 0.06 + 0.0025)
(1 + X)2 = (1.05)(1.0625)
(1 + X)2 = (1.115625)
1 + X = 1.1156251/2
X = 1.1156251/2 – 1
• X = 0.05623 = 5.623%
Market Segmentation Theory

• Under the Market Segmentation Theory, rates are determined

by supply and demand conditions in each maturity range of the
yield curve. For example, 30-year rates are influenced by the
supply and demand for mortgages.
• According to this theory, investors have a strong preference for
a specific maturity and would require a premium to invest in
other bonds with different maturities. As a result, there can be a
risk premium for different maturities along the yield curve, but
they do not necessarily increase with maturity as they do under
the Liquidity Premium Theory. As a result, the yield curve may
not rise or fall continuously.
Key Bond Characteristics

Each bond can be characterized by several factors. These include:

• Face Value
• Coupon Rate
• Coupon
• Maturity
• Call Provisions
• Put Provisions
• Sinking Fund Provisions
Call Provisions
• Many bonds contain a provision that enables the issuer to buy
the bond back from the bondholder at a pre-specified price prior
to maturity. This price is known as the call price.
• A bond containing a call provision is said to be callable. This
provision enables issuers to reduce their interest costs if rates
fall after a bond is issued, since existing bonds can then be
replaced with lower yielding bonds. Since a call provision is
disadvantageous to the bond holder, the bond will offer a higher
yield than an otherwise identical bond with no call provision.
• A call provision is known as an embedded option, since it can’t
be bought or sold separately from the bond.
Put Provisions
Some bonds contain a provision that enables the buyer to sell the
bond back to the issuer at a pre-specified price prior to maturity.
This price is known as the put price. A bond containing such a
provision is said to be putable. This provision enables bond
holders to benefit from rising interest rates since the bond can be
sold and the proceeds reinvested at a higher yield than the
original bond. Since a put provision is advantageous to the bond
holder, the bond will offer a lower yield than an otherwise
identical bond with no put provision.
Sinking Fund Provisions
Some bonds are issued with a provision that requires the issuer to
repurchase a fixed percentage of the outstanding bonds each year,
regardless of the level of interest rates. A sinking fund reduces the
possibility of default; default occurs when a bond issuer is unable
to make promised payments in a timely manner. Since a sinking
fund reduces credit risk to bond holders, these bonds can be
offered with a lower yield than an otherwise identical bond with
no sinking fund.
Bond Pricing
A bond’s price equals the present value of its expected future
cash flows. The rate of interest used to discount the bond’s cash
flows is known as the yield to maturity (YTM.)
 :

C=Periodic Coupon Payment

Y=Yield to Maturity
Pricing T=Time period till bonds
maturity date
Coupon t=time
Bonds
This formula shows that the price of a bond is the present value of
its promised cash flows. As an example, suppose that a bond has
a face value of $1,000, a coupon rate of 4% and a maturity of four
years. The bond makes annual coupon payments. If the yield to
maturity is 4%, the bond’s price is determined as follows:
These results show the following important relationship:
• if Y > coupon rate, P < face value
• if Y = coupon rate, P = face value
• if Y < coupon rate, P > face value
These results also demonstrate that there is an inverse
relationship between yields and bond prices:
• when yields rise, bond prices fall
• when yields fall, bond prices rise
Repo Rates

The interest rate at which the central bank in a country

repurchases government securities (such as Treasury Securities)
from Commercial
Banks. The central bank raises the repo rate when it wishes to
reduce the money supply in the
short, while it lowers the rate when it wishes to increase
the money supply and stimulate growth
 24-03-2021
Bonds and Derivatives
Key Bond Characteristics

Each bond can be characterized by several factors. These include:

• Face Value
• Coupon Rate
• Coupon
• Maturity
• Call Provisions
• Put Provisions
• Sinking Fund Provisions
Call Provisions
• Many bonds contain a provision that enables the issuer to buy
the bond back from the bondholder at a pre-specified price prior
to maturity. This price is known as the call price.
• A bond containing a call provision is said to be callable. This
provision enables issuers to reduce their interest costs if rates
fall after a bond is issued, since existing bonds can then be
replaced with lower yielding bonds. Since a call provision is
disadvantageous to the bond holder, the bond will offer a higher
yield than an otherwise identical bond with no call provision.
• A call provision is known as an embedded option, since it can’t
be bought or sold separately from the bond.
Put Provisions
Some bonds contain a provision that enables the buyer to sell the
bond back to the issuer at a pre-specified price prior to maturity.
This price is known as the put price. A bond containing such a
provision is said to be putable. This provision enables bond
holders to benefit from rising interest rates since the bond can be
sold and the proceeds reinvested at a higher yield than the
original bond. Since a put provision is advantageous to the bond
holder, the bond will offer a lower yield than an otherwise
identical bond with no put provision.
Sinking Fund Provisions
Some bonds are issued with a provision that requires the issuer to
repurchase a fixed percentage of the outstanding bonds each year,
regardless of the level of interest rates. A sinking fund reduces the
possibility of default; default occurs when a bond issuer is unable
to make promised payments in a timely manner. Since a sinking
fund reduces credit risk to bond holders, these bonds can be
offered with a lower yield than an otherwise identical bond with
no sinking fund.
Bond Pricing
A bond’s price equals the present value of its expected future
cash flows. The rate of interest used to discount the bond’s cash
flows is known as the yield to maturity (YTM.)
 :

C=Periodic Coupon Payment

Y=Yield to Maturity
Pricing T=Time period till bonds
maturity date
Coupon t=time
Bonds
This formula shows that the price of a bond is the present value of
its promised cash flows. As an example, suppose that a bond has
a face value of $1,000, a coupon rate of 4% and a maturity of four
years. The bond makes annual coupon payments. If the yield to
maturity is 4%, the bond’s price is determined as follows:
These results show the following important relationship:
• if Y > coupon rate, P < face value
• if Y = coupon rate, P = face value
• if Y < coupon rate, P > face value
These results also demonstrate that there is an inverse
relationship between yields and bond prices:
• when yields rise, bond prices fall
• when yields fall, bond prices rise
Adjusting for Semi-Annual Coupons
For a bond that makes semi-annual coupon payments, the
following adjustments must be made to the pricing formula:
• the coupon payment is cut in half
• the yield is cut in half
• the number of periods is doubled
As an example, suppose that a bond has a face value of $1,000, a
coupon rate of 8% and a maturity of two years. The bond makes
semi-annual coupon payments, and the yield to maturity is
6%. The semi-annual coupon is $40, the semi-annual yield is 3%,
and the number of semi-annual periods is four. The bond’s price
is determined as follows:

= 38.83 + 37.70 + 36.61 + 924.03 = $1,037.17

• As an alternative to this pricing
formula, a bond may be priced by
treating the coupons as an
annuity; the price is therefore
equal to the present value of an
annuity (the coupons) plus the
present value of a sum (the face
value.) This method of valuing
bonds will use the formula:
The bond in the previous example can be priced using
this alternate bond valuation formula as follows:

= 148.68 + 888.49 = $1,037.17

 A Zero-coupon bond does not
make any coupon payments;
Pricing Zero instead, it is sold to investors at
a discount from face value. The
Coupon difference between the price
paid for the bond and the face
Bonds value, known as a capital gain,
is the return to the investor.
The pricing formula for a zero
coupon bond is:
F=Face Value
y= YTM
T=No of periods till maturity
As an example, suppose that a one-year zero-coupon bond is
issued with a face value of $1,000. The discount rate for this
bond is 8%. What is the market price of this bond?
In order to be consistent
with coupon-bearing
bonds, where coupons are
typically made on a semi-
annual basis, the yield will
be divided by 2, and the
number of periods will be
multiplied by 2:
Yield Measures

There are different types of yield measures that may be used to

represent the approximate return to a bond. These include:
• yield to maturity (YTM)
• yield to call (YTC)
• current yield
Yield to Maturity (YTM)
The discount rate used in the bond pricing formula is also known as
the bond’s yield to maturity (YTM) or yield. This equals the rate of
return earned by a bond holder (known as the holding period return)
if:
• The bond is held to maturity
• The coupon payments are reinvested at the yield to maturity
• A bond’s YTM is the unique discount rate at which the market price
of the bond equals the present value of the bond’s cash flows:
Market Price = PV (Cash Flows)
The yield to maturity of a bond can be determined from the
bond’s market price, maturity, coupon rate and face value. As an
example, suppose that a bond has a face value of $1,000 and will
mature in ten years. The annual coupon rate is 5%; the bond
makes semi-annual coupon payments. With a price of $950,
what is the bond’s yield to maturity?
• It is impossible to solve for the yield to maturity algebraically;
instead, this must be done using a financial calculator or
Microsoft Excel. For example, a bond’s yield to maturity can be
computed in Excel using the RATE function:
= RATE(nper, pmt, pv, [fv], [type], [guess])
where:
• nper = number of periods
• pmt = periodic payment
• pv = present value
• fv = future value
• type = 0 for ordinary annuity
• 1 for annuity due
• guess = initial guess
• The variables in brackets (fv, type and guess) are optional
values; the value of type is set to zero if it is not specified. Guess
can be used to provide an initial estimate of the rate, which
could potentially speed up the calculation time.
• Note that either pv or fv must be negative, and the other must
be positive. The negative value is considered to be a cash
outflow, and the positive value is considered to be a cash inflow.
• Also note that entering semi-annual periods and coupon
payments will produce a semi-annual yield; in order to convert
this into an annual yield (on a bond-equivalent basis), the semi-
annual yield is doubled.
• In this example,
• nper = 20
• pmt = $25
• pv = $950
• fv = $1,000
At a price of $950, the semi-annual yield to maturity is:
=RATE(nper, pmt, pv, [fv], [type], [guess])
= RATE(20, 25, -950, 1000)
= 2.83%
The annual yield is (2.83%)(2) = 5.66%
At a price of $1000, the semi-annual yield to maturity is:
=RATE(nper, pmt, pv, [fv], [type], [guess])
= RATE(20, 25, -1000, 1000)
= 2.50%
The annual yield is (2.50%)(2) = 5.00%
At a price of $1050, the semi-annual yield to maturity is:
=RATE(nper, pmt, pv, [fv], [type], [guess])
= RATE(20, 25, -1050, 1000)
= 2.19%
The annual yield is (2.19%)(2) = 4.38%
Yield to Call (YTC)
For a bond that is callable, the yield to call may be used as a
measure of return instead of the yield to maturity. The process is
similar to computing yield to maturity, except that the maturity
date of the bond is replaced with the next call date. This is
because yield to call is based on the assumption that the bond will
be called on the next call date. The face value is replaced with
the call price since this is the amount that the investor will
receive if the bond is called.
• As an example, suppose that a ten-year bond was issued two
years ago and is callable in three years at a price of $1,100. The
bond’s face value is $1,000 and its coupon rate is 7%. Coupons
are paid on an annual basis; the current market price of the
bond is $1,200. What is the yield to call?
In this case, the bond will mature in eight years, but it can be
called in three years. If the bond is called, the investor will
receive a price of $1,100 instead of the face value of $1,000. The
yield to call is computed as follows:
The yield to call is:
=RATE(nper, pmt, pv, [fv], [type], [guess])
= RATE(3, 70, -1200, 1100)
= 3.14%
Current Yield
• The current yield is simpler measure of the rate of return to a
bond than the yield to maturity. Current yield is measured as
the ratio of the bond’s annual coupon payment to the bond’s
market price.
As an example, suppose that a bond was issued with a coupon
rate of 8% and a face value of $1,000. The bond has a current
market price of $900. The current yield is computed as:
$80/$900 = 8.89%
This measure has the benefit of simplicity. It suffers from the
drawback that it does not account for the time value of money.
Repo Rates

The interest rate at which the central bank in a country

repurchases government securities (such as Treasury Securities)
from Commercial
Banks. The central bank raises the repo rate when it wishes to
reduce the money supply in the
short, while it lowers the rate when it wishes to increase
the money supply and stimulate growth
Excel’s Specialized Bond Functions
PRICE
The Excel function price is implemented as follows:
= PRICE(settlement, maturity, rate, yld, redemption, frequency, [basis])
where:
settlement = date on which the bond owner pays for the bond
maturity = maturity date of the bond; this is the date on which the owner receives
the principal
rate = annual coupon rate; this is the percentage of face value that is paid to the
bond owner each year
yld = the bond’s yield to maturity
redemption = face value (as a percentage of 100)
frequency = the number of coupon payments per year; annually = 1, semi-annually
=2
basis = the day-count convention
Options for basis are
• 0 = 30/360
• 1 = actual/actual
• 2 = actual/360
• 3 = actual/365
• 4 = European 30/360
Treasury bonds follow the actual/actual convention. Corporate
and municipal bonds follow the 30/360 convention. Money
market instruments (e.g., Treasury bills, commercial paper, etc.)
follow the actual/360 convention.
Note that the settlement date and maturity date are represented
as numerical values in Excel. The date January 1, 1900 is
represented as 1; all later dates represent the number of days that
have passed since January 1, 1900.
As an example, suppose that a bond is sold on June 15, 2016 with
a maturity date of June 15, 2036. The coupon rate is 8%, the
yield is 9%, the face value is $1,000 and the bond makes semi-
annual coupon payments. Also assume that the bond uses a
30/360 day count convention for computing coupon
payments. What is the price of the bond?
This is computed with the PRICE function as follows:
= PRICE(settlement, maturity, rate, yld, redemption, frequency,
[basis])
= PRICE(42536, 49841, 8%, 9%, 100, 2, 0)
= $907.99
YIELD
The Excel function yield is implemented as follows:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
where:
pr = price (as a percentage of face value)
As an example, for the bond in the previous example, the yield
can be determined as follows:
= YIELD(settlement, maturity, rate, pr, redemption, frequency,
[basis])
= YIELD(42536, 49841, 8%, 90.799, 100, 2, 0)
= 9.00%
Note that the price of the bond is entered as 90.799 instead of
907.99; this indicates that the price is 90.799 percent of the face
value. Also note that 42536 represents June 15, 2016 and 49841
represents 49841.
These numerical values can be obtained with the Excel function
DATEVALUE. For example,
= DATEVALUE(“6/15/16”)
= 42536
Managing Bond Portfolio & Risks
Measures of Interest Rate Risk& Bond
Portfolio Management
The management of bond portfolios or fixed income portfolios
introduces several unique challenges; among the most important is the
ability to determine the risk associated with fixed income
instruments. Three of the most important measures of interest rate
risk are known as:
• Macaulay duration
• Modified duration
• Convexity
These measures are widely used to determine how sensitive a bond’s
price is to changes in market yields or in other words: what is the risk
associated with this bond portfolio when market rates change?
These measures also called interest rate risk measures quantify
the bond portfolio risk. These interest rate risk measures have
one major drawback: they are based on the assumption that all
promised cash flows will actually materialize. This assumption is
especially problematic when bonds contain embedded options,
which can be used to alter the pattern of cash flows from a
bond. In order to properly account for embedded options, more
advanced risk measures have been developed; these are known
as:
• Effective duration
• Effective convexity
The above interest rate risk measures only help you quantify
understand the level of risk in a bond portfolio. They do not help
you mitigate or manage the bond portfolio risk! Different types of
strategies can be used to manage the returns and risk of a bond
portfolio. Two of the more widely-used bond portfolio risk
management strategies are:
• Indexing
• Immunization
Interest Rate Risk
Macaulay Duration
Duration is a measure of the sensitivity of a bond’s price to
changes in the yield to maturity. Duration enables an investor to
directly compare the risks of bonds with different face values,
maturities, coupons, etc. The concept of duration was first
defined by Frederic Macaulay in 1938 in his book “The Movement
of Interest Rates, Bond Yields and Stock Prices in the United
States Since 1856.” As a result, the form of duration that he
suggested is known as Macaulay duration. Since the development
of Macaulay duration, a closely related measure known as
modified duration has become widely used in many applications.
Macaulay duration is the sum of the present values of a bond’s
time-weighted cash flows divided by the bond’s price. In the
special case of a zero coupon bond, Macaulay duration equals the
bond’s maturity.
The formula for computing Macaulay
Duration is:

where:
DMAC = Macaulay Duration
t = a time index
n = the number of periods until the
bond’s maturity date
CFt = the bond’s cash flow at time t
y = the bond’s yield to maturity
P = the bond’s price
• As an example, suppose that a ten-year U.S. Treasury note that
was issued seven years ago with a coupon rate of 6% and a face
value of $1,000. Also assume that newly-issued (“on-the-run”)
three-year Treasury notes offer a coupon rate of 3%, and that
the note pays semi-annual coupons.
• The price of the bond is computed as the present value of the
bond’s future cash flows. The bond in this example makes a
semi-annual coupon payment of $30 twice per year; at maturity,
the face value (principal) of $1,000 is repaid. Each cash flow is
discounted by the semi-annual yield to maturity, which is one-
half of 3%, or 1.5%.
• Since the bond makes semi-annual coupon payments, this result
is measured in terms of semi-annual periods. The Macaulay
Duration expressed in years is 5.60/2 = 2.80. This indicates
that it takes 2.8 years before the present value of the cash flows
adds up to 1,000, the initial price of the bond.
Modified Duration

Although Macaulay Duration is a useful measure of interest rate

risk, for many applications the interpretation is not
convenient. An alternative form of duration was developed that
can be interpreted as the sensitivity of a bond’s price to a change
in the yield curve. This version of duration is known as modified
duration
Modified duration is computed as follow:
where:
• D = modified duration
• P = the bond’s price
• dP = an instantaneous change in the bond’s
price
• dy = an instantaneous change in the bond’s
yield
Price-Yield Curve
This can be approximated as:
where:
DeltaP = a small change in the bond’s price
Delta y = a small change in the bond’s yield
The ratio deltaP/deltaY is known as
the first derivative of the price with respect
to the yield. This can be thought of as
the slope of the price-yield curve, which
shows the relationship between the price of
a bond and the yield.
The price-yield curve is negatively sloped; in other words, the
first derivative is negative throughout. The price-yield curve is
also convex to the origin; this means that the curve is “bowed”,
and the second derivative is positive throughout.
Modified duration equals the negative of the slope of the price-
yield curve, divided by the bond’s price. The negative sign
ensures that modified duration will be a positive number, which
is easier to interpret than a negative number. The slope is scaled
by the price to correct for the fact that not all bonds are issued
with the same face value and can therefore have very different
prices. The goal of modified duration is to measure
the sensitivity of the bond’s price to changes in the yield curve.
Modified duration can be computed by using calculus; this is
accomplished by differentiating the bond’s pricing function (the
present value of the future cash flows) with respect to the bond’s
yield, and then multiplying the result by –(1/P).
s a simpler alternative, the modified
duration can be computed directly
from the bond’s Macaulay duration
with the following equation:

where: m = the coupon frequency

In this formula, m = 1 for a bond that
pays annual coupons, and m = 2 for a
bond that pays semi-annual coupons.
y = the bond’s yield to maturity
Based on the previous example, the modified duration
is computed as follows:

This can be interpreted as follows. If the bond’s yield

rises by 1%, the bond’s price will fall by approximately
2.76%, and vice versa. Note that the smaller the change
in yield, the more accurate this estimate will be.
Factors That Affect Modified Duration

Modified duration is a function of a bond’s maturity and coupon

rate.
• Duration is an increasing function of maturity, since a longer
maturity bond has more cash flows that are affected by a given
change in yield.
• Duration is a decreasing function of the coupon rate. Zero
coupon bonds are most heavily affected by changes in yield
since all cash flows take place on a single date; with coupon-
bearing bonds, the impact of changes in yield is spread out
among the coupon payments.
As an example, the following table shows the modified duration
of four bonds: a 5 year zero coupon bond, a 5 year 5% coupon
bond, a 10 year zero coupon bond and a 10 year 5% coupon
bond. The yield curve is flat at 4% (i.e., yield is 4% for all
maturities.) Coupons are assumed to be paid semi-annually.
 MODIFIED The chart shows that
BOND the 5 year zero coupon
DURATION bond has a modified
duration of 4.90, which
5 year 0% coupon 4.90 is well below the 9.80
modified duration of
the 10 year zero
5 year 5% coupon 4.41 coupon
bond. Similarly, the 5
year 5% coupon bond
10 year 0% coupon 9.80 has a modified
duration of 4.41, while
the 10 year 5% coupon
10 year 5% coupon 7.92 bond has a modified
duration of 7.92.
In both cases, the 10 year bond has more cash flows than the five
year bond, and some of these cash flows will take place further in
the future. Therefore, a given change in yield will have a greater
impact on the present value of the 10 year bond’s cash flows; as a
result, its price will be more sensitive to changes in yields than
the 5 year bond.
• The chart also shows that the 5 year zero coupon bond has a
modified duration of 4.90, which is greater than the 4.41
modified duration of the 5 year 5% coupon bond. Similarly, the
10 year zero coupon bond has a modified duration of 9.80
compared with a modified duration of 7.92 for the 10 year 5%
coupon bond.
• In both cases, the zero coupon bond has a higher duration than
the 5% coupon bond. This is because with a zero coupon bond,
all cash flows take place at maturity; as a result, a given change
in yield has a greater impact on the present value of the cash
flows than it does for a bond with a higher coupon and the same
maturity.
Using Modified Duration to Estimate Changes in Bond Prices

Based on the formula for computing

modified duration, the approximate change
in the price of a bond may be estimated from
the bond’s modified duration, price and the
change in yield. Since

The change in a bond’s price due to a given

change in yield can be determined by
rearranging this equation algebraically:
As an example, suppose that for the ten-year U.S. Treasury note that
was issued seven years ago with a coupon rate of 6% and a face value
of $1,000, the yield rises from 3% to 3.01%. What will be the impact
on the price of the bond? The change in yield is computed as:
Δy = 0.0301 – 0.0300 = 0.0001
This is 0.01%, which is also known as a basis point. One hundred
basis points equal one percent. Therefore, the change in the bond
price will be:
ΔP = -(2.76)(1,085.46)(0.0001)
= -$0.30
The price will fall by about thirty cents as a result of an increase of one
basis point in the bond’s yield. The new price will be approximately
$1,085.46 – $0.30 = $1,085.16.
In order to determine the percentage change in
a bond’s price due to a given change in yield,
the following equation can be used:

The change in P divided by the original price of

P gives the percentage change in the price.
Based on the previous example:

The price of the bond falls by about

0.0276% as a result of an increase in
the yield by one basis point.
Convexity

A bond’s convexity refers to the sensitivity of the bond’s modified

duration to changes in yield. Based on the price-yield curve:
• At low yields, the modified duration changes very quickly as the
yield changes (since the price-yield curve is very steep);
therefore, the convexity is high
• At higher yields, the modified duration changes very slowly as
the yield changes (since the price-yield curve is relatively flat);
therefore, the convexity is low
The Math
As an example, suppose that the price of a bond is initially P0, and then
changes to P1. In this case, the first difference of P equals DP = P1 –
P0. Suppose the bond price now changes from P1 to P2. There have
now been two changes: DP1 = P1 – P0 and DP2 = P2 – P1. The second
difference of P equals:

Δ2P = ΔP2 – ΔP1

= (P2 – P1) – (P1 – P0)

= P2 – 2P1 + P0
Convexity can be computed by using calculus; this is
accomplished by computing the second derivative of the bond’s
price function with respect to the bond’s yield, and then
multiplying the result by 1/P.
As a simpler alternative, the convexity can be
computed with the following equation:
where:
C = convexity
t = a time index
n = the number of periods until the bond’s maturity
date
CFt = the bond’s cash flow at time t
y = the bond’s yield to maturity
P = the bond’s price
Using the example of the ten-year U.S. Treasury note that was
issued seven years ago with a coupon rate of 6%, a face value of
$1,000 and a yield of 3%, the convexity can be computed as
follows.
= 57.38 + 169.59 + 334.17 + 548.73 + 810.92 + 38,402.38
= $40,323.18
When this is multiplied by 1/P, the convexity is determined to be:
C = 40,323.18 / 1,085.45 = 37.15
Convexity is positive for option-free bonds. Convexity can be
negative if a bond contains an embedded call option. An
embedded call option enables the issuer to repurchase the bond
at a fixed price (known as the call price) at a specified time in the
future. With very low yields, the likelihood of a bond being called
is extremely high; as a result, investors will not pay more than the
call price for a bond, no matter how low yields are. This results in
a phenomenon known as “price compression”, in which the price
of a bond is prevented from rising above the call price while the
price can still fall due to rising yields. When yields are extremely
low, the bond’s convexity can become negative as the price curve
flattens out.
Using both modified duration and convexity to estimate the change in a bond’s price due
to a given change in yield can be estimated more precisely than with modified duration
alone; the equation is:
• For the ten-year U.S. Treasury note that was issued seven years
ago with a coupon rate of 6%, a face value of $1,000 and a yield
of 3%, the change in the price due to an increase in the yield to
3.01% can be computed as follows:
• ΔP = -2.76(1,085.46)(0.0001) + 0.5(37.15)(1,085.46)(0.0001)2
• = -0.30 + 0.0002
• = -$0.2998
• n this case, using convexity in addition to modified duration
added very little accuracy to the estimated price change. In
other cases, using convexity can make a significant
improvement in the estimation of the price change. This can
occur when the convexity is very large; unlike modified
duration, which cannot exceed the maturity of the bond, the
convexity of a bond can reach a value in the thousands.
In addition to computing the change in
a bond’s price due to a given change in
yield, modified duration and convexity
can be combined to give the
approximate percentage change in the
price of a bond, as follows:
Effective Duration and Convexity
One of the major drawbacks of the modified duration and
convexity measures is that they are based on the assumption that
all promised cash flows to a bond will actually be made. This is
not necessarily the case for a bond containing an embedded
option. In particular, with an embedded call option, the bond can
be repurchased prior to maturity by the issuer. This will result in
the last few years’ worth of cash flows not being received by the
bond owner. As a result, the modified duration and convexity
measures may not accurately reflect the risk of this bond.
In order to correct for this problem, two alternative measures
were developed: effective duration and effective convexity. With
these measures, a model of the term structure of interest rates is
needed. Within the model, yields are increased by a specified
number of basis points, and the impact on the price is
observed. Yields are then decreased by the same number of basis
points, and the impact on the price is observed.
Effective duration is defined as
follows:

where:
DE = effective duration
P0 = the initial price of a bond
P– = the price of the bond after
the yield curve shifts down by
Δy basis points
P+ = the price of the bond after
the yield curve shifts up by Δy
basis points
Dy = the change in yield
For example, suppose that a
bond’s price is currently
$1,050.00. If the yields in a
term structure model are
reduced by 25 basis points,
the bond’s price rises by
$26.25 to $1,076.25. If the
yields are increased by 25
basis points, the bond’s
price falls by $25.89 to
$1,024.11. In this case, the
effective duration would be:

where:
CE = effective convexity
Passive Bond Portfolio Strategies
Different types of strategies can be used to manage the returns
and risk of a bond portfolio; some of the more widely-used
strategies are known as:
• Indexing
• Immunization
• With an indexing strategy, the portfolio manager attempts to
replicate a bond index, such as the Standard and Poor’s 500
Bond Index. Due to the varied features of bonds, this is a more
difficult objective to accomplish than replicating a stock market
index.
With an immunization strategy, the risk of a portfolio is managed
by attempting to ensure that the duration of a bond portfolio
matches a specified investment time horizon. The goal is to
ensure that the portfolio will provide a certain rate of return by
the end of the time horizon
Summary of what we learnt on Bonds
• TVM • Bond Pricing
• Future Value of a Sum • Bond Characteristics
• Annuity • Yield Measures
• YTM
• Ordinary Annuity • YTC
• Future Value • Current Yield
• Annuinities Due • Repo Rates
• Term Structure • Excel Sheet functions
• Yield Curve • Managing Bond Portfolio & Risks
• Interest Rate Risk
• Interest Rates • Macaulay Duration
• Discount Factors • Modified Duration
• Convexity
• Yields • Measures of Risk
• Expectations theory • Effective Duration
• Effective Convexity
• Liquidity Premium Theory • Risk Mitigation
• Indexing
• Immunization
Primary instruments
• Shares
• Bonds
• Currency
• Commodity
Others
• Futures
• Options
• Swaps
Secondary
• Derivatives
 31032021
Derivatives , Futures, Options & Swaps
• The best movie to
understand the 2008
crash….
Derivative(Hull)
A derivative is defined as a financial instrument whose value depends
on(or derives from) the values of other more basic
underlying.variables,Very often variables underlying derivatives are
traded assets.A stock option for example is a derivative whose value is
dependent on the price of the stock..However derivatives can be
dependent on any variable-commodity, oil, currency, bonds, interest
rates
Derivatives( Ashwani Gujral)
Derivatives as the name suggests are financial instruments whos value
is dependent on the underlying asset.The underlying security in the
case of equity derivatives is a equity share.A share of equity can only
provide unhedged position whether long or short and the entire risk of
the transaction lies with the trader or investor. Whereas derivatives
allow futures, options , swaps and other means for speculation and
hedging
Forward
A relatively simple derivative is a forward contract.It is an agreement to
buy or sell an asset at a certain future time at a certain price.A forward
contract is usually contracted between two financial institutions.A
forward contract is very popular in the foreign currency market to
hedge foreign currency risk.
A forward contract is a customized contract between two parties
to buy or sell an asset at a specified price on a future date. A
forward contract can be used for hedging or speculation,
although its non-standardized nature makes it particularly apt
for hedging
• A forward contract is a customizable derivative contract between two
parties to buy or sell an asset at a specified price on a future date.
• Forward contracts can be tailored to a specific commodity, amount,
and delivery date.
• Forward contracts do not trade on a centralized exchange and are
considered over-the-counter (OTC) instruments.
• For example, forward contracts can help producers and users of
agricultural products hedge against a change in the price of an
underlying asset or commodity.
• Financial institutions that initiate forward contracts are exposed to a
greater degree of settlement and default risk compared to contracts
that are marked-to-market regularly.
• Unlike standard futures contracts, a forward contract can be
customized to a commodity, amount, and delivery date. Commodities
traded can be grains, precious metals, natural gas, oil, or even
poultry. A forward contract settlement can occur on a cash or delivery
basis.
• Forward contracts do not trade on a centralized exchange and are
therefore regarded as over-the-counter (OTC) instruments. While
their OTC nature makes it easier to customize terms, the lack of a
centralized clearinghouse also gives rise to a higher degree
of default risk.
• Because of their potential for default risk and lack of a centralized
clearinghouse, forward contracts are not as easily available to retail
investors as futures contracts
SpotContract
• Is a contract to buy or sell an asset almost immediately
• In finance, a spot contract, spot transaction, or simply spot, is
a contract of buying or selling a commodity, security or
currency for immediate settlement (payment and delivery)
on the spot date, which is normally two business days after the
trade date.
• T+n
• Settlement is on the nth day, 3rd day usually
• There is a relationship between interest rates and forward
prices and spot
Long Positions
• The term long position describes what an investor has
purchased when they buy a security or derivative with the
expectation that it will rise in value.
• Being long is a measurement of time
• In forward contracts the party who assumes a long position
agrees to buy the underlying asset on a certain specified future
date for a certain specified price
• This term will be used for bullish acts
Short Position
• In a forward contract the party that agrees to sell the asset on a
specific date for a specific price is said to have taken a short position
• A short, or a short position, is created when a trader sells a
security first with the intention of repurchasing it or covering it
later at a lower price. A trader may decide to short a security
when she believes that the price of that security is likely to
decrease in the near future. There are two types of short
positions: naked and covered. A naked short is when a trader
sells a security without having possession of it.
• In general bearish action
Payoff
K= Delivery Price as per contract at time T
ST=Spot Price at time T

For
Long= ST-K
Short=K-ST
Consider the following example of a forward contract. Assume
that an agricultural producer has two million bushels of corn to
sell six months from now and is concerned about a potential
decline in the price of corn. It thus enters into a forward contract
with its financial institution to sell two million bushels of corn at a
price of $4.30 per bushel in six months, with settlement on a cash
basis
Long or Short? Who?
In six months, the spot priceof corn has three possibilities:
1.It is exactly $4.30 per bushel. In this case, no monies are
owed by the producer or financial institution to each other and
the contract is closed.
2.It is higher than the contract price, say $5 per bushel. The
producer owes the institution $1.4 million, or the difference
between the current spot price and the contracted rate of $4.30.
3.It is lower than the contract price, say $3.50 per
bushel. The financial institution will pay the producer $1.6
million, or the difference between the contracted rate of $4.30
and the current spot price.
More practical…
Consider a stock that is worth 60$ and pays no dividend.You can
borrow or lend money at 5%, what should the forward price of the
stock be?
Assuming no rerating of the stock the answer is 60$ +5%=63$
If the stock is expected to touch 67$, borrow 60$ buy one share and
sell it at 67$.After paying the loan on 3$ you have a profit of 4$
Longs
If the stock were expected to touch 58$, investor investor owning this
can sell it and enter into a forward contract for 1 year at 58$.The
proceeds of 60$ earn 3$ and he owns the stock back at a profit of 2 $.
Net 5$
Shorts
Futures
What is a future?
Like a forward contract, a future is an agreement to buy an asset at a
certain time at the future for a certain price.U
Unlike forward contracts future contracts are traded on a exchange
Tomake trading possible the exchange specificies certain standardized
features of the contract.
As the two parties do not know each other the exchange also provides
a mechanism that gives the two parties a gaurantee that the contract
will be honoured
Clearinghouse

Because futures contracts are standardized, there is an active

market in which participants can trade their futures contracts
before their expiry date. Such an exchange is called a
clearinghouse. New York Mercantile Exchange (NYMEX),
Chicago Board of Trade (CBOT) and Chicago Board Options
Exchange (CBOE) are the main exchanges on which futures can
be traded.
Margin requirement

• Parties looking to purchase or sell futures contracts are required

to maintain a margin with the exchange. It gives the exchange
an assurance that they have necessary funds to honor their
obligation in event of any adverse price movement.
• Initial margin refers to the amount that the parties deposit with
the clearinghouse at the inception of the futures contract. If as a
result of the marking to market process, the party’s balance
decreases below the maintenance margin, the minimum margin
that they are required to maintain, they receive a margin call.
• Margins, sometimes set as a percentage of the value of the futures contract,
must be maintained throughout the life of the contract to guarantee the
agreement, as over this time the price of the contract can vary as a function of
supply and demand, causing one side of the exchange to lose money at the
expense of the other.
• To mitigate the risk of default, the product is marked to market on a daily basis
where the difference between the initial agreed-upon price and the actual daily
futures price is re-evaluated daily. This is sometimes known as the variation
margin, where the futures exchange will draw money out of the losing party's
margin account and put it into that of the other party, ensuring the correct loss or
profit is reflected daily. If the margin account goes below a certain value set by
the exchange, then a margin call is made and the account owner must replenish
the margin account.
• On the delivery date, the amount exchanged is not the specified price on the
contract but the spot value since any gain or loss has already been previously
settled by marking to market.
Various types of margin:
-Clearing margin
-Customer margin
-Initial margin
-Maintenance margin
-Margin/Equity ratio
-Return on Margin
Marking to market

• Marking to market refers to the process adopted by

clearinghouses/exchanges to calculate and settle the net payoff
on futures contracts periodically, typically daily. The exchange
credits the differential amount in the margin account if a party
gains on the futures contract and draws on the margin balance
if the party has lost money due to price movement.
Forward price formula

The futures price i.e. the price at which the buyer commits to
purchase the underlying asset can be calculated using the
following formulas:
FP0 = S0 × (1+i)t
Where,
FP0 is the futures price,
S0 is the spot price of the underlying,
i is the risk-free rate and t is the time period.
Value of a futures contract
The formula is a little different for futures contract in which the
underlying asset has cash inflows or outflows during the term of
the futures contract, for example stocks, bonds, commodities,
etc.
The value of a futures contract is different from the future price. It
is the value of the long or short position in the futures contract
itself and it depends on whether the spot price of the underlying
asset at the time of valuation is higher or lower than the agreed
futures price and the risk-free interest rate.
Value at inception

The value of futures contract for the buyer (i.e. the party with long
position) at inception is zero and the value at expiration equals
the difference between the associated spot rate at the expiration
date minus the futures price i.e. the price at which the futures
contract was purchased.
• VT = ST − F0
• Where,
VT is the value at expiration,
ST is the spot price at expiration, and
F0 is the futures price looked through the futures contract.
Value after inception but before expiry date
• Because futures contracts are traded on an exchange, parties
might sell them any time between the inception date and the
expiration date. In such an event, the value of the futures
contract equals the difference between the spot price at that
time (denoted as St) minus the present value of the futures price
locked at time t. The value of futures future F0 that we expected
to get at time T can be worked out by discounting the futures
price (F0) at risk-free rate r for the remaining time period (i.e. T
minus t). The value of the futures contract can be worked out as
follows:
Vt= St - F0/(1+r)*(T-t)
• Stock futures offer a variety of usages to the investors. Some of the key usages
are mentioned below:
Investors can take long term view on the underlying stock using stock futures.
• Stock futures offer high leverage. This means that one can take large position
with less capital. For example, paying 20% initial margin one can take position for
100 i.e. 5 times the cash outflow.
Futures may look overpriced or underpriced compared to the spot and can offer
opportunities to arbitrage or earn risk-less profit. Single stock futures
offer arbitrage opportunity between stock futures and the underlying cash
market. It also provides arbitrage opportunity between synthetic futures (created
through options) and single stock futures.
When used efficiently, single-stock futures can be an effective risk management
tool. For instance, an investor with position in cash segment can minimize either
market risk or price risk of the underlying stock by taking reverse position in an
appropriate futures contract.
• Presently, stock futures are settled in cash. The final settlement
price is the closing price of the underlying stock.
The investor can square up his position at any time till the
expiry. The investor can first buy and then sell stock futures to
square up or can first sell and then buy stock futures to square
up his position. E.g. a long (buy) position in December Infy
futures, can be squared up by selling December Infy futures.
• The outstanding positions in stock futures are marked-to-
market daily. The closing price of the respective futures contract
is considered for marking to market. The notional loss / profit
arising out of mark to market is paid / received on T+1 basis.
• The profits and losses would depend upon the difference
between the price at which the position is opened and the price
at which it is closed. Let an investor have a long position of one
November Stock "A" Futures @ 430. If the investor square up
his position by selling November Stock "A" futures @ 450, the
profit would be Rs. 20 per share. In case, the investor squares
up his position by selling November Stock "A" futures @ 400,
the loss would be Rs. 30 per share.
• According to L.C.Gupta Committee Report on Derivatives, at the
time of introduction of Derivatives Contracts on any underlying the
value of the contract should be at least Rs. 2 lakhs. This value of Rs.
2 lakhs is divided by the market price of the individual stock to arrive
at the initial 'market lot' for it. It may be mentioned here that the only
exception to this rule is the 'mini' contract on the S&P BSE
SENSEX (both futures and Options).

Similarly, you can enter an order for Sell Nov Dec stating the
difference you want to receive. This would mean that you are selling
a December Contract and buying a November Contract and
receiving the difference.
• One can trade in spread contracts on the Derivative Segment of
BSE. Spreads are the contracts for differential price. This
means that in case you want to buy a December contract and
sell November contract, you can enter an order for Buy Nov
Dec stating the difference you want to pay. This would mean
that you are buying a December Contract and selling a
November contract.
Deposit upfront the initial margin
Similarly, you can enter an order for Sell Nov Dec stating the
difference you want to receive. This would mean that you are
selling a December Contract and buying a November Contract
and receiving the difference.
Pair Trading
• This trading strategy involves taking a position on the relative
performance of two stocks. It is achieved by buying futures on
the stock expected to perform well and selling futures on the
stock anticipated to perform poorly. The overall gain or loss
depends on the relative performance of the two stocks.

Similarly it is possible to take a position in the relative

performance of a stock versus a market index. For example,
traders who would like to take only company specific risk could
buy/sell the relative index future.
india
• Indian futures settlements take place on the last Thursday of every
month the current
• Current month’s futures expire on the last Thursday of every month
• If a trader has to carry his position to the next month he has to shift
his position to the next month’s future
• Presently in India t would be liquidhe current month’s stock or indicex
futures the only one that can be traded.Futures for succeeding
months are not liquid enough to be traded.
• For example during April only April futures will be liquid enough.The
May and June may open up after last Thursday of April
options
Options are financial instruments that are derivatives based on
the value of underlying securities such as stocks. An options
contract offers the buyer the opportunity to buy or sell—
depending on the type of contract they hold—the underlying
asset. Unlike futures the holder is not required to buy or sell the
asset if they choose not to.
• Call options allow the holder to buy the asset at a stated price
within a specific timeframe
• Put options allow the holder to sell the asset at a stated price
within a specific timeframe
• Each option contract will have a specific expiration date by
which the holder must exercise their option. The stated price on
an option is known as the strike price. Options are typically
bought and sold through online or retail brokers
Options are a versatile financial product. These contracts involve
a buyer and a seller, where the buyer pays an
options premium for the rights granted by the contract. Each call
option has a bullish buyer and a bearish seller, while put options
have a bearish buyer and a bullish seller.
• Options Contracts usually represent 100 shares of the
underlying security, and the buyer will pay a premium fee for
each contract. For example, if an option has a premium of 35
cents per contract, buying one option would cost $35 ($0.35 x
100 = $35). The premium is partially based on the strike price—
the price for buying or selling the security until the expiration
date. Another factor in the premium price is the expiration date.
Just like with that carton of milk in the refrigerator, the expiration
date indicates the day the option contract must be used. The
underlying asset will determine the use-by date. For stocks, it is
usually the last Thursday of the contract's month.
Traders and investors will buy and sell options for several
reasons. Options speculation allows a trader to hold a leveraged
position in an asset at a lower cost than buying shares of the
asset. Investors will use options to hedge or reduce the risk
exposure of their portfolio. In some cases, the option holder
can generate income when they buy call options or become an
options writer. Options are also one of the most direct ways
to invest in oil. For options traders, an option's daily
trading volume and open interest are the two key numbers to
watch in order to make the most well-informed investment
decisions
American options can be exercised any time before the
expiration date of the option, while European options can only be
exercised on the expiration date or the exercise date. Exercising
means utilizing the right to buy or sell the underlying
security.India is the European model
Options Risk Metrics: The Greeks
• The “Greeks" is a term used in the options market to describe
the different dimensions of risk involved in taking an options
position, either in a particular option or a portfolio of options.
These variables are called Greeks because they are typically
associated with Greek symbols. Each risk variable is a result of
an imperfect assumption or relationship of the option with
another underlying variable. Traders use different Greek values,
such as delta, theta, and others, to assess options risk and
manage option portfolios.
Delta
Delta(Δ) represents the rate of change between the option's price
and a $1 change in the undelying asset’s price. In other words,
the price sensitivity of the option relative to the underlying. Delta
of a call option has a range between zero and one, while the
delta of a put option has a range between zero and negative one.
For example, assume an investor is long a call option with a delta
of 0.50. Therefore, if the underlying stock increases by $1, the
option's price would theoretically increase by 50 cents.
• For options traders, delta also represents the hedge ratio for
creating a delta neutral position. For example if you purchase a
standard American call option with a 0.40 delta, you will need to
sell 40 shares of stock to be fully hedged. Net delta for a
portfolio of options can also be used to obtain the portfolio's
hedge ration.
• A less common use of an option’s delta is it's current probability
that it will expire in-the money. For instance, a 0.40 delta call
option today has an implied 40% probability of finishing in-the-
money.
Theta
• Theta (Θ) represents the rate of change between the option price
and time, or time sensitivity - sometimes known as an option's time
decay. Theta indicates the amount an option's price would decrease
as the time to expiration decreases, all else equal. For example,
assume an investor is long an option with a theta of -0.50. The
option's price would decrease by 50 cents every day that passes, all
else being equal. If three trading days pass, the option's value would
theoretically decrease by $1.50.
• Theta increases when options are at-the-money, and decreases
when options are in- and out-of-the money. Options closer to
expiration also have accelerating time decay. Long calls and long
puts will usually have negative Theta; short calls and short puts will
have positive Theta. By comparison, an instrument whose value is
not eroded by time, such as a stock, would have zero Theta.
Gamma
• Gamma (Γ) represents the rate of change between an
option’s delta and the underlying asset's price. This is called
second-order (second-derivative) price sensitivity. Gamma
indicates the amount the delta would change given a $1 move
in the underlying security. For example, assume an investor is
long one call option on hypothetical stock XYZ. The call option
has a delta of 0.50 and a gamma of 0.10. Therefore, if stock
XYZ increases or decreases by $1, the call option's delta would
increase or decrease by 0.10.
• Gamma is used to determine how stable an option's delta is: higher
gamma values indicate that delta could change dramatically in
response to even small movements in the underlying's price.Gamma
is higher for options that are at-the-money and lower for options that
are in- and out-of-the-money, and accelerates in magnitude as
expiration approaches. Gamma values are generally smaller the
further away from the date of expiration; options with longer
expirations are less sensitive to delta changes. As expiration
approaches, gamma values are typically larger, as price changes
have more impact on gamma.
• Options traders may opt to not only hedge delta but also gamma in
order to be delta-gamma-neutral, meaning that as the underlying
price moves, the delta will remain close to zero.
Vega

• Vega (V) represents the rate of change between an option's

value and the underlying asset’s implied volatility . This is the
option's sensitivity to volatility. Vega indicates the amount an
option's price changes given a 1% change in implied volatility.
For example, an option with a Vega of 0.10 indicates the
option's value is expected to change by 10 cents if the implied
volatility changes by 1%.
• Because increased volatility implies that the underlying
instrument is more likely to experience extreme values, a rise in
volatility will correspondingly increase the value of an option.
Conversely, a decrease in volatility will negatively affect the
value of the option. Vega is at its maximum for at-the-money
options that have longer times until expiration.
Rho
• Rho (p) represents the rate of change between an option's
value and a 1% change in the interest rate. This measures
sensitivity to the interest rate. For example, assume a call
option has a rho of 0.05 and a price of $1.25. If interest rates
rise by 1%, the value of the call option would increase to $1.30,
all else being equal. The opposite is true for put options. Rho is
greatest for at-the-money options with long times until
expiration.
• Minor Greeks
• Some other Greeks, with aren't discussed as often,
are lambda,epsilon, vomma, vera, speed, zomma, color, ultima.
• These Greeks are second- or third-derivatives of the pricing
model and affect things such as the change in delta with a
change in volatility and so on. They are increasingly used in
options trading strategies as computer software can quickly
compute and account for these complex and sometimes
esoteric risk factors.
 Risk and Profits From Buying Call
Options
• As mentioned earlier, the call options let the holder buy an underlying security at the
stated strike price by the expiration date called the expiry. The holder has no
obligation to buy the asset if they do not want to purchase the asset. The risk to the
call option buyer is limited to the premium paid. Fluctuations of the underlying stock
have no impact.
• Call options buyers are bullish on a stock and believe the share price will rise above
the strike price before the option's expiry. If the investor's bullish outlook is realized
and the stock price increases above the strike price, the investor can exercise the
option, buy the stock at the strike price, and immediately sell the stock at the current
market price for a profit.
• Their profit on this trade is the market share price less the strike share price plus the
expense of the option—the premium and any brokerage commission to place the
orders. The result would be multiplied by the number of option contracts purchased,
then multiplied by 100—assuming each contract represents 100 shares.
• However, if the underlying stock price does not move above the strike price by the
expiration date, the option expires worthlessly. The holder is not required to buy the
shares but will lose the premium paid for the call.
Risk and Profits From Selling Call
Options
• Selling call options is known as writing a contract. The writer receives the
premium fee. In other words, an option buyer will pay the premium to the writer—
or seller—of an option. The maximum profit is the premium received when selling
the option. An investor who sells a call option is bearish and believes the
underlying stock's price will fall or remain relatively close to the option's strike
price during the life of the option.
• If the prevailing market share price is at or below the strike price by expiry, the
option expires worthlessly for the call buyer. The option seller pockets the
premium as their profit. The option is not exercised because the option buyer
would not buy the stock at the strike price higher than or equal to the prevailing
market price.
• However, if the market share price is more than the strike price at expiry, the
seller of the option must sell the shares to an option buyer at that lower strike
price. In other words, the seller must either sell shares from their portfolio
holdings or buy the stock at the prevailing market price to sell to the call option
buyer. The contract writer incurs a loss. How large of a loss depends on the cost
basis of the shares they must use to cover the option order, plus any brokerage
order expenses, but less any premium they received.
• As you can see, the risk to the call writers is far greater than the risk exposure of
call buyers. The call buyer only loses the premium. The writer faces infinite risk
because the stock price could continue to rise increasing losses significantly.
Risk and Profits From Buying Put Options
• Put options are investments where the buyer believes the underlying stock's
market price will fall below the strike price on or before the expiration date of the
option. Once again, the holder can sell shares without the obligation to sell at the
stated strike per share price by the stated date
• Since buyers of put options want the stock price to decrease, the put option is
profitable when the underlying stock's price is below the strike price. If the
prevailing market price is less than the strike price at expiry, the investor can
exercise the put. They will sell shares at the option's higher strike price. Should
they wish to replace their holding of these shares they may buy them on the open
market.
• Their profit on this trade is the strike price less the current market price, plus
expenses—the premium and any brokerage commission to place the orders. The
result would be multiplied by the number of option contracts purchased, then
multiplied by 100—assuming each contract represents 100 shares.
• The value of holding a put option will increase as the underlying stock price
decreases. Conversely, the value of the put option declines as the stock price
increases. The risk of buying put options is limited to the loss of the premium if
the option expires worthlessly.
Risk and Profits From Selling Put Options
• Selling put options is also known as writing a contract. A put option writer believes the underlying
stock's price will stay the same or increase over the life of the option—making them bullish on the
shares. Here, the option buyer has the right to make the seller, buy shares of the underlying asset
at the strike price on expiry.
• If the underlying stock's price closes above the strike price by the expiration date, the put option
expires worthlessly. The writer's maximum profit is the premium. The option isn't exercised
because the option buyer would not sell the stock at the lower strike share price when the market
price is more.
• However, if the stock's market value falls below the option strike price, the put option writer is
obligated to buy shares of the underlying stock at the strike price. In other words, the put option
will be exercised by the option buyer. The buyer will sell their shares at the strike price since it is
higher than the stock's market value.
• The risk for the put option writer happens when the market's price falls below the strike price.
Now, at expiration, the seller is forced to purchase shares at the strike price. Depending on how
much the shares have depreciated, the put writer's loss can be significant.
• The put writer—the seller—can either hold on to the shares and hope the stock price rises back
above the purchase price or sell the shares and take the loss. However, any loss is offset
somewhat by the premium received.
• Sometimes an investor will write put options at a strike price that is where they see the shares
being a good value and would be willing to buy at that price. When the price falls, and the option
buyer exercises their option, they get the stock at the price they want, with the added benefit of
receiving the option premium.
• Suppose that Microsoft shares are trading at $108 per share
and you believe that they are going to increase in value. You
decide to buy a call option to benefit from an increase in the
stock's price.
• You purchase one call option with a strike price of $115 for one
month in the future for 37 cents per contact. Your total cash
outlay is $37 for the position, plus fees and commissions (0.37 x
100 = $37).
• If the stock rises to $116, your option will be worth $1, since you
could exercise the option to acquire the stock for $115 per share and
immediately resell it for $116 per share. The profit on the option
position would be 170.3% since you paid 37 cents and earned $1—
that's much higher than the 7.4% increase in the underlying stock
price from $108 to $116 at the time of expiry.
• In other words, the profit in dollar terms would be a net of 63 cents or
$63 since one option contract represents 100 shares [($1 - 0.37) x
100 = $63].
• If the stock fell to $100, your option would expire worthlessly, and
you would be out $37 premium. The upside is that you didn't buy 100
shares at $108, which would have resulted in an $8 per share, or
$800, total loss. As you can see, options can help limit your
downside risk.
Options Spreads

• Options spreads are strategies that use various combinations of

buying and selling different options for a desired risk-return
profile. Spreads are constructed using vanilla options, and can
take advantage of various scenarios such as high- or low-
volatility environments, up- or down-moves, or anything in-
between.
• Spread strategies, can be characterized by their payoff or
visualizations of their profit-loss profile, such as bull call
spreads /iron condors
•i
Swaps
• Derivatives contracts can be divided into two general families:
1. Contingent claims (e.g., options)
2. Forward claims, which include exchange-traded futures,
forward contracts, and swaps
A swap is an agreement between two parties to exchange
sequences of cash flows for a set period of time. Usually, at the
time the contract is initiated, at least one of these series of cash
flows is determined by a random or uncertain variable, such as
an interest rate, foreign exchange rate, equity price, or
commodity price
• Conceptually, one may view a swap as either a portfolio of
forward contracts or as a long position in one bond coupled with
a short position in another bond. The two most common and
most basic types of swaps: interest rate and currency
• In finance, a swap is a derivative contract in which one party
exchanges or swaps the values or cash flows of one asset for
another.
• Of the two cash flows, one value is fixed and one is variable
and based on an index price, interest rate, or currency
exchange rate.
• Swaps are customized contracts traded in the over-the-counter
(OTC) market privately, versus options and futures traded on a
public exchange.
• The plain vanilla interest rate and currency swaps are the two
most common and basic types of swaps.
• Unlike most standardized options and futures contracts, swaps
are not exchange-traded instruments. Instead, swaps are
customized contracts that are traded in the over-the-counter
(OTC) market between private parties
• Firms and financial institutions dominate the swaps market, with
few (if any) individuals ever participating. Because swaps occur
on the OTC market, there is always the risk of
a counterparty defaulting on the swap.
• The first interest rate swap occurred between IBM and the
World Bank in 1981.1 However, despite their relative youth,
swaps have exploded in popularity. In 1987, the International
Swaps and Derivatives Association reported that the swaps
market had a total notional value of $865.6 billion.2 By mid-
2006, this figure exceeded $250 trillion, according to the Bank
for International Settlements.3 That's more than 15 times the
size of the U.S. public equities market.
Plain Vanilla Interest Rate Swap

The most common and simplest swap is a plain vanilla interest

rate swap. In this swap, Party A agrees to pay Party B a
predetermined, fixed rate of interest on a notional principal on
specific dates for a specified period of time. Concurrently, Party B
agrees to make payments based on a floating interest rate to
Party A on that same notional principal on the same specified
dates for the same specified time period. In a plain vanilla swap,
the two cash flows are paid in the same currency. The specified
payment dates are called settlement dates, and the times
between are called settlement periods. Because swaps are
customized contracts, interest payments may be made annually,
quarterly, monthly, or at any other interval determined by the
parties.
• For example, imagine ABC Co. has just issued $1 million in five-
year bonds with a variable annual interest rate defined as
the LIBOR plus 1.3% (or 130 basis points). Also, assume that
LIBOR is at 2.5% and ABC management is anxious about an
interest rate rise.
• The management team finds another company, XYZ Inc., that is
willing to pay ABC an annual rate of LIBOR​ plus 1.3% on a
notional principal of $1 million for five years. In other words,
XYZ will fund ABC's interest payments on its latest bond
issue. In exchange, ABC pays XYZ a fixed annual rate of 5% on
a notional value of $1 million for five years. ABC benefits from
the swap if rates rise significantly over the next five years. XYZ
benefits if rates fall, stay flat, or rise only gradually.
• LIBOR, or London Interbank Offered rate is the interest rate
offered by London banks on deposits made by other banks in
the Eurodollar markets. The market for interest rate swaps
frequently (but not always) used LIBOR as the base for the
floating rate until 2020. The transition from LIBOR to other
benchmarks, such as the secured overnight financing
rate (SOFR), began in 2020.
Scenario 1
If LIBOR rises by 0.75% per year, Company ABC's total interest
payments to its bondholders over the five-year period amount to
$225,000. Let's break down the calculation:
 Libor + 1.30% Variable Interest 5% Interest Paid ABC's Gain XYZ's Loss
Paid by XYZ to by ABC to XYZ
ABC
Year 1 3.80% $38,000 $50,000 -$12,000 $12,000
Year 2 4.55% $45,500 $50,000 -$4,500 $4,500
Year 3 5.30% $53,000 $50,000 $3,000 -$3,000
Year 4 6.05% $60,500 $50,000 $10,500 -$10,500
Year 5 6.80% $68,000 $50,000 $18,000 -$18,000
Total $15,000 ($15,000)
• In this scenario, ABC did well because its interest rate was fixed
at 5% through the swap. ABC paid $15,000 less than it would
have with the variable rate. XYZ's forecast was incorrect, and
the company lost $15,000 through the swap because rates rose
faster than it had expected.
• If LIBOR rises 0,25% per year
Scenario 2
In the second scenario, LIBOR rises by 0.25% per year:

Variable
5% Interest
Interest Paid
Libor + 1.30% Paid by ABC ABC's Gain XYZ's Loss
by XYZ to
to XYZ
ABC
Year 1 3.80% $38,000 $50,000 ($12,000) $12,000
Year 2 4.05% $40,500 $50,000 ($9,500) $9,500
Year 3 4.30% $43,000 $50,000 ($7,000) $7,000
Year 4 4.55% $45,500 $50,000 ($4,500) $4,500
Year 5 4.80% $48,000 $50,000 ($2,000) $2,000
Total ($35,000) $35,000
• In this case, ABC would have been better off by not engaging in the
swap because interest rates rose slowly. XYZ profited $35,000 by
engaging in the swap because its forecast was correct.
• This example does not account for the other benefits ABC might
have received by engaging in the swap. For example, perhaps the
company needed another loan, but lenders were unwilling to do that
unless the interest obligations on its other bonds were fixed.
• In most cases, the two parties would act through a bank or other
intermediary, which would take a cut of the swap. Whether it is
advantageous for two entities to enter into an interest rate swap
depends on their comparative advantage in fixed or floating-
rate lending markets.
• The instruments exchanged in a swap do not have to be
interest payments. Countless varieties of exotic swap
agreements exist, but relatively common arrangements include
commodity swaps, currency swaps, debt swaps, and total return
swaps.
Commodity Swaps

Commodity swaps involve the exchange of a floating commodity

price, such as the Brent Crude oil spot price, for a set price over
an agreed-upon period. As this example suggests, commodity
swaps most commonly involve crude oil.
• In a currency swap, the parties exchange interest and principal
payments on debt denominated in different currencies. Unlike
an interest rate swap, the principal is not a notional amount, but
it is exchanged along with interest obligations. Currency swaps
can take place between countries. For example, China has
used swaps with Argentina, helping the latter stabilize its foreign
reserves.The US Federal Reserve engaged in an aggressive
swap strategy with European central banks during the 2010
European financial crisis to stabilize the euro, which was falling
in value due to the Greek debt crisis
Debt-Equity Swap
• A debt-equity swap involves the exchange of debt for equity—in
the case of a publicly-traded company, this would mean bonds
for stocks. It is a way for companies to refinance their debt or
reallocate their capital structure.
Total Return Swap
• In a total return swap, the total return from an asset is
exchanged for a fixed interest rate. This gives the party paying
the fixed-rate exposure to the underlying asset—a stock or
an index. For example, an investor could pay a fixed rate to one
party in return for the capital appreciation
plus dividend payments of a pool of stocks.
Credit Default Swap
• A (CDS) consists of an agreement by one party to pay the lost
principal and interest of a loan to the CDS buyer if a
borrower defaults on a loan. Excessive leverage and poor risk
management in the CDS market were contributing causes of
the 2008 financial crisis
Swaps Summary

• A financial swap is a derivative contract where one party

exchanges or "swaps" the cash flows or value of one asset for
another. For example, a company paying a variable rate of
interest may swap its interest payments with another company
that will then pay the first company a fixed rate. Swaps can also
be used to exchange other kinds of value or risk like the
potential for a credit default in a bond.
 14042021
Transition, Black Scholes, CAPM
Black Scholes
Black Scholes model
• The Black-Scholes model, also known as the Black-Scholes-
Merton (BSM) model, is a mathematical model for pricing an
options contract. In particular, the model estimates the variation
over time of financial instruments.
• The Black-Scholes model is one of the most important concepts
in modern financial theory. It was developed in 1973 by Fischer
Black, Robert Merton, and Myron Scholes and is still widely
used today. It is regarded as one of the best ways of
determining the fair price of options. The Black-Scholes model
requires five input variables: the strike price of an option, the
current stock price, the time to expiration, the risk-free rate, and
the volatility.
• Also called Black-Scholes-Merton (BSM), it was the first widely
used model for option pricing. It's used to calculate the
theoretical value of options using current stock prices, expected
dividends, the option's strike price, expected interest rates, time
to expiration and expected volatility.
• The initial equation was introduced in Black and Scholes' 1973
paper, "The Pricing of Options and Corporate Liabilities,"
published in the Journal of Political Economy.
• Black passed away two years before Scholes and Merton were
awarded the 1997 Nobel Prize in economics for their work in
finding a new method to determine the value of derivatives (the
Nobel Prize is not given posthumously; however, the Nobel
committee acknowledged Black's role in the Black-Scholes
model).
Black-Scholes posits that instruments, such as stock shares or
futures contracts, will have a lognormal distribution of prices
following a random walk with constant drift and volatility. Using
this assumption and factoring in other important variables, the
equation derives the price of a European-style call option.
The inputs for the Black-Scholes equation are volatality the price
of the underlying asset, the strike price of the option, the time
until expiration of the option, and the risk-free interest rate. With
these variables, it is theoretically possible for options sellers to
set rational prices for the options that they are selling.
Furthermore, the model predicts that the price of heavily traded
assets follows a geometric Brownian motion with constant drift
and volatility. When applied to a stock option, the model
incorporates the constant price variation of the stock, the time
value of money, the option's strike price, and the time to the
option's expiry.
 The first known work applying advanced
mathematics to option pricing (and to
finance in general) was by French
Louis Bachelier mathematician Louis Bachelier. His thesis,
titled Theory of Speculation [1], used the
(1900 concept now known as Brownian motion
(from physics) or Wiener process (from
mathematics) to model stock option prices –
the same concept that provides the
foundation of Black-Scholes and many other
financial models. The date of Bachelier’s
thesis defense, 29 March 1900, is
sometimes mentioned as the origin of
quantitative finance.
Like many other revolutionary ideas,
Bachelier’s work received rather mixed
reaction at that time . He got a pass, though
not the highest grade for his thesis, and had
it published in a prestigious journal.
Nevertheless, his ideas, which he continued
to develop in subsequent years, didn’t have
much influence in finance until many
decades later.
 Paul Samuelson

-First American to win the

Nobel Prize for Economics
-The best text book
-Combined Math and
Economics
-MIT Prof
• Edward Thorp-Prof of Math
• Sheen Kassouf-Prof of Economics
• Univrsity of California, Irvine
Milestone Check
• Fundamentals over
• Trading
• Mainly Statistics
• GARCH, ARIMA, Bayesian logic, Candlesticks etc, variance, filters
• Backtesting
• R/python
210421
Pioneers of Algorithmic Trading

The beginning of computational finance dates back to 1930s

when some investors started calculating mathematics formulas to
price stocks and bonds. In 1950s Harry Markowitz introduced
computational finance in order to solve the portfolio selection
problem. One of the major problems was the shortage of
computer power at the time what made analysis difficult.
Therefore, the mathematical science diverged by making
simplifying assumptions to establish patterns in forms that did not
require sophisticated computer science analysis.
• Further, in the 1960s, hedge fund managers Ed Thorp and Michael
Goodkin in collaboration with Harry Markowitz were the first to ever use
computers for arbitrage trading. The introduction of a personal computer in
the late 1970s and early 1980s brought about an exploration of wide range
of computational finance applications. Although, many of the new
techniques emerged from signal processing, rather than the traditional
field of computational economics e.g optimization and time series analysis.
• As a result of the cold war by the end of the 1980s, a large number of
physicists and applied mathematicians moved into the study of finance
forming new groups of financial engineer and quantitative portfolio
managers who started to develop technologies for automated trading as
we know it today.
• First technologies were, of course, primitive because also the computers
were not yet developed, however, they presented the foundations to the
technologies we know and use today.
Who is a Trader?

• In simple words, any individual who buys and sells Financial

assets in any financial market is a trader. This individual or
trader can trade on the behalf of any other person as well here.
A trader is usually someone who trades in shorter time periods
as compared to an investor. This simply means that a trader
holds assets for a short period to make profits on short-term
trends. Whereas, an investor tends to hold assets for a longer-
term.
Now, who is a Quant/Quantitative Analyst?

• A quantitative analyst is the one who designs a complex framework

for financial institutions that aids them to price and trade securities in
the financial market. Quants can be of two types:
• Front office quants - These are the ones who directly provide the
trader with the price of the financial securities or the trading tools.
• Back office quants - These quants are there to validate the
framework and create new strategies after conducting thorough
research.
• Moving ahead, now let us find out more about algorithmic trading
and its association with Maths.
• Usually, when quants work, they keep an eye on the performance of the
market.
• But the interesting part is: “How do they predict or forecast on the basis of
market data?”
• And the answer is: They do it with MATHS!
• Digging deeper, in this process, data is bought from the stock market and
is analysed. It is then on the basis of this data that they come up with the
possible percentage of odds (say, 65% or 75% and so on) with regard to
the movements of stock prices. This is known as “predicting/forecasting
the possibility of the stock prices in the long term or short term”.
• Those involved in creating algorithms for High-Frequency Trading
(HFT) keep in mind the involvement of a large number of trades in a short
period.
When and How Mathematics made it to
Trading: A historical tour
• Now, it was not until the late sixties that mathematicians made
their first entry into the financial world of Trading. It all started
with a professor of mathematics called Edward Thorp, at the
University of California, who published a book called Beat the
Market in 1967. In this book, he claimed that he had provided
the foolproof way of earning money on the stock market. Also,
this method/way was entirely based on a system that he had
devised for beating casinos at blackjack. It is said that it
became extremely famous due to which the casinos were
forced to change their rules to “Beat the Market”.
• Specifically, Beat the Market concept was nothing but the
process of selling the stocks and bonds at one price and then
buying them back at a lower price. This strategy became so
popular and efficient that Edward Thorp founded a hedge fund
named as Princeton/Newport Partners. This hedge fund
proceeded to rule over the markets and hence, it became a full-
fledged strategy. Soon after, a generation of physicists entered
the depressed job market but on observing the quantum of
money that could be made on Wall Street, many of them moved
into finance consequently.
• It was also observed that in Britain, the fall of the Soviet Union
brought an influx of Warsaw Pact scientists. Hence, they brought
with themselves a new methodology based on the concept of
“analysing the data” along with the understanding that sufficient
computer firepower can help predict the market. This brought along
a new concept of quantitative analysis and a maths genius named
Jim Simons became famous in bringing enough knowledge in the
particular sphere. In 1982, Jim Simons also founded an exceptional
hedge fund management company called Renaissance
Technologies.
• All in all, this was the brief on “how mathematics took off in
algorithmic trading” and is so successful. Now let us head to the
Mathematical concepts for algorithmic trading which are the core of
this article.
Mathematical Concepts

• Starting with the mathematical concepts of trading, it is a must

to mention that mathematical concepts play an important role in
algorithmic trading. Let us take a look at the broad categories of
different concepts here:
• Descriptive Statistics
• Probability Theory
• Linear Algebra
• Linear Regression
• Calculus
Descriptive Statistics
• Let us walk through descriptive statistics, which summarize a
given data set with brief descriptive coefficients. These can be a
representation of either the whole or a sample from the
population.
Measure of Central Tendency
• Here, Mean, Median and Mode are the basic measures of
central tendency. These are quite useful when it comes to
taking out average value from a data set consisting of various
values. Let us understand each measure one by one:
Mean

This one is the most used concept in the various fields concerning
mathematics and in simple words, it is the average of the given dataset.
Thus, if we take five numbers in a data set, say, 12, 13, 6, 7, 19, 21, the
formula of the mean is

which makes it :
(12 + 13 + 6 + 7 + 19 + 21)/6 = 13
Furthermore, the trader tries to initiate the trade on the basis of the mean
(moving average) or moving average crossovers
Here, let us understand two types of moving averages based on the
ranges (number of days) of the time period they are calculated in and the
moving average crossover:
Faster moving average (Shorter time
period) -
• A faster moving average is the mean of a data set (stock prices)
calculated over a short period of time, say past 20 days.
Slower moving average (Longer time
period)
• A slower moving average is the one that is the mean of a data
set (stock prices) calculated from a longer time period say 50
days.
• Now, a faster-moving average and a slower moving average
also come to a position together where a “crossover” occurs.
• According to Wikipedia, “A crossover occurs when a faster-
moving average (i.e., a shorter period moving average) crosses
a slower moving average (i.e. a longer period moving average).
In other words, this is when the shorter period moving average
line crosses a longer period moving average line.”
• Here, to explain it better, the graph image above is showing
three moving lines. Blue one shows the trend line of the stock
prices in general. It is further disintegrated into green and
orange lines. The green one indicates a slower-moving average
and orange one indicates a faster-moving average.
• Now starting with the green line, (slower moving average) the
entire trend line shows the varying means of stock prices over
longer time periods. The trend line follows a zig-zag pattern and
there are different crossovers.
• For example, there is a crossover between October, 2018 and
January, 2019 where orange line (faster-moving average)
comes from above and crosses the green one (slower-moving
average) while going down. This indicates that any individual or
firm would be selling the stocks at this point since it shows a
slump in the market.
• This crossover point is called the “meeting point”. After the
meeting point, ahead both the lines go down and then go up
after a point to create one more (and then other) crossover(s).
• Now, it is very important to note here that the “meeting point” is
considered bullish if the faster-moving average crosses over the
slower-moving average and goes beyond in the upward
direction. On the contrary, it is considered bearish if the faster-
moving average drops below the slower-moving average and
goes beyond down. This is so because in the former scenario, it
shows that in a short time, there came an upward trend for
particular stocks. Whereas, in the latter scenario it shows that in
the past few days there was a downward trend.
• For example, we will be taking the same instances of the 20-
days' moving average for faster-moving average and 50 days'
moving average for slower-moving average. If 20-days' moving
average goes up and crosses 50- days' moving average, it will
show a bullish market since it indicates an upward trend in the
past 20-days’ stocks. Whereas, if the 20-days' moving average
goes below the 50-days' moving average, it will
be bearish since it means that the stocks fell in the past 20-
days.
• In short, Mean is a statistical indicator used for estimating a
company’s or even the market’s stock performance over a
period of time. This period of time can be days, months and
even years.
Going forward, mean can also be
computed with the help of an
excel sheet, with the following
formula:
=Average(B2: B6)

Let us understand what we have

done in the image above. The
image shows the stock cap of
different companies belonging to
an industry over a period of time
(can be days, months, or years).
• Now, to get the moving average (mean) of this industry in this
particular time period, we need the formula =(Average(B2: B6))
to be applied against “Mean stock price”. This formula gives the
command to the excel to average out the stock prices of all the
companies mentioned from row B2 to B6.
• As we apply this formula and press “Enter” we get the result
330. This is one of the simplest methods to compute Mean. Let
us see how to compute the same in python code ahead.
import yfinance as yf aapl = yf.download('AAPL','2018-12-26', '2019-12-26')

• For further use, in all the concepts, let us assume values

on the basis of Apple’s (AAPL) data set. In order to keep it
universal, we have taken the daily stock price data of Apple, Inc.
from Dec 26, 2018, to Dec 26, 2019. You can download
historical data from Yahoo Finance.
• Now, For downloading the Apple closing price data, we will
use the following for all python code based calculations
ahead:
import yfinance as yf
aapl = yf.download('AAPL','2018-12-26', '2019-12-26')
mean = np.mean (aapl[‘Adj Close’]) print(mean)

• In python, for taking out the mean of closing prices, the code
will be as follows:
mean = np.mean (aapl[‘Adj Close’])
print(mean)

RESULT???
Median
• Sometimes, the data set values can have a few values which
are at the extreme ends, and this might cause the mean of the
data set to portray an incorrect picture. Thus, we use the
median, which gives the middle value of the sorted data set.
• To find the median, you have to arrange the numbers in
ascending order and then find the middle value. If the dataset
contains an even number of values, you take the mean of the
middle two values.
• For example, if the list of numbers are: 12, 13, 6, 7, 19, then,
• In ascending order, the numbers are: 6, 7, 12, 13, 19
• Now, we know there are in total 5 numbers and the formula for
Median is:
• (n+1)/2 value.
• Hence, it will be n = 5 and
• (5+1)/2 value will be 6/2= 3rd value.
• Here, the 3rd value in the list is 12.
• So, the median becomes 12 here.
• Mainly, the advantage of the median is that unlike the mean, it
remains extremely valid in case of extreme values of data set
which is the case in stocks.
• Median is required in case the average is to be calculated from
a large data set, in which, the median shows an average which
is a better representation of the data set.
• For example, in case the data set is given as follows with values in
INR:
• 75,000, 82,500, 60,000, 50,000, 1,00,000, 70,000 and 90,000.
• Calculation of the median needs the prices to be first placed in
ascending order, thus, prices in ascending order are:
• 50,000, 60,000, 70,000, 75,000, 82,500, 90,000, 1,00,000
• Now, the calculation of the median will be:
• As there are 7 items, the median is (7+1)/2 item, which makes it the
4th item. The 4th item in the ascending order is INR 75,000.
• As you can see, INR 75,000 is a good representation of the data set,
so this will be an ideal one.
• Excel sheet helps in the
following way to compute
median:
• =Median(B2:B6)
•
median = np.median (aapl[‘Adj Close’]) print(median)

• The python code here will be:

median = np.median (aapl[‘Adj Close’])

print(median)
Mode
• Mode is a very simple concept since it takes into consideration
that number in the data set which is repetitive and occurs the
most. Also, the mode is known as a modal value, representing
the highest count of occurrences in the group of a data. It is
also interesting to note that like mean and median, a mode is a
value that represents the whole data set. It is extremely
imperative to note that, in some of the cases there is a
possibility of there being more than one mode in a given data
set. And that data set which has two modes will be known as
bimodal.
• Similar to Mean and Median,
Mode can also be calculated
in the excel sheet as shown in
the image above. For
example, you can put in the
values of different companies
in the excel sheet and take
out the Mode with the formula
=Mode.SNGL(B1: B5)
• (B1: B5) - represents the
values from cell B1 till B5
import statistics
mode = statistics.mode (aapl[‘Adj Close’])

Mode in Python

Now, if we take the closing prices prices of Apple from Dec 26,
2018, to Dec 26, 2019, we will find there is no repeating value,
and hence the mode of closing prices does not exist.
• Hence, the mode does not make sense while observing closing
price values.
• Coming to the significance of the mode, it is most helpful when
you need to take out the repetitive stock price from the previous
particular time period. This time period can be days, months
and even years. Basically, the mode of the data will help you
understand if the same stock price is expected to repeat in the
future or not.
• Also, the mode is best utilised when you want to plot histograms
and visualize the frequency distribution.
Measure of Dispersion
• It displays how scattered the data is around the central point. It
simply tells the variation of each data value from one another,
which helps to give a representation of the distribution of the
data. Also, it portrays the homogeneity and heterogeneity of the
distribution of the observations.
• In short, it simply shows how much the entire data varies from
their average value.
Measure of dispersion can be divided into:
• Range
• Quartile Deviation
• Mean Absolute Data or Mean Deviation
• Variance
• Standard Deviation
Now, let us understand the concept of each category.
Range

• This is the most simple out of all the measures of dispersion and is also easy to
understand. Range simply implies the difference between two extreme observations or
numbers of the data set.
• For example, let X max and X min be two extreme observations or numbers. Here, Range
will be the difference between the two of them.
• Hence,
• Range = X max - X min
• It is also very important to note that Quant analysts keep a close follow up on ranges. This
happens because the ranges determine the entry as well as exit points of trades. Not only
the trades, but Range also helps the traders and investors in keeping a check on trading
periods.
• This makes the investors and traders indulge in Range-bound Trading strategies, which
simply imply following a particular trendline. The trendlines are formed by high priced
stocks (following an upper trendline) and low priced stocks (following a lower trendline). In
this, the trader can purchase the security at the lower trendline and sell it at a higher
trendline to earn profits.
Quartile Deviation:

• This is the type which divides a data set into quarters. It consists of First Quartile
as Q1, Second Quartile as Q2 and Third Quartile as Q3.
• Here,
• Q1 - is the number that comes between the smallest and the median of the data
(1/4th) or top 25%
• Q2 - is the median of the data or
• Q3 - is the number that comes between the median of data and the largest
number (3/4th) or lower 25%
• n - is the total number of values
• And the formula for Quartile deviation is Q = ½ * (Q3 - Q1)
• Since,
• Q1 is top 25%, the formula for Q1 is - ¼ (n+1)
• Q3 is also 25%, but the lower one, so the formula is - ¾ (n+1)
• Hence, Quartile deviation = ½ * [(¾ (n+1) - ¼ (n+1)]
• The major advantage, as well as the disadvantage of using this
formula, is that it uses half of the data to show the dispersion from
the mean or average.
• You can use this type of measure of dispersion for studying the
dispersion of the observations that lie in the middle.
• This type of measures of dispersion helps you understand dispersion
from the observed value and hence, differentiates between the large
values in different Quarters.
• In the financial world, when you have to study a large data set (stock
prices) in different time periods and want to understand the
dispersed value (prices) from an observed one (average-median),
Quartile deviation can be used.
Mean Absolute Deviation:

• This type of dispersion is the arithmetic meaof the deviations

between the numbers in a given data set from their mean or
median (average).
• Hence,
• The formula of Mean Absolute Deviation is:
• (D0 + D1 + D2 + D3 + D4 ….Dn)/ n
• Here,
• n = Total number of deviations in the data set and
D0, D1, D2, D3 are the deviations of each value from
the average or median or mean in the data set and Dn
means the end value in the data set.
Explaining the Mean deviation, we will take a look at
the image below, which shows a “computed mean” of a
data set and the difference between each value (in the
dataset) from the mean value. These differences or the
deviations are shown as D0, D1, D2, and D3, …..D7.
For an instance, if the mean values are as follows:
• Then, the Mean here will be calculated using the mean formula:
• 3 + 6 + 6 + 7 + 8 + 11 + 15 + 16 / 8 = 9
• As the mean comes out to be 9, next step is to find the
deviation of each data value from the Mean value. So, let us
compute the deviations, or let us subtract 9 from each value to
find D0, D1, D2, D3, D4, D5, D6, D7, and D8, which gives us
the values as such:
As we are now clear about all the deviations, let us see the mean
value and all the deviations in the form of an image to get even
more clarity on the same:
hen, the Mean here will be
calculated using the mean
formula:
3 + 6 + 6 + 7 + 8 + 11 + 15 + 16 /
8=9
As the mean comes out to be 9,
next step is to find the deviation of
each data value from the Mean
value. So, let us compute the
deviations, or let us subtract 9
from each value to find D0, D1,
D2, D3, D4, D5, D6, D7, and D8,
which gives us the values as such:
As we are now clear about all the deviations, let us see the
mean value and all the deviations in the form of an image to
get even more clarity on the same:

Hence, from a large data set, the mean deviation represents

the required values from observed data value accurately.
• It is important to note that Mean deviation helps with a large
dataset with various values which is especially the case in the
stock market.
Going ahead, Variance is a related concept and is further
explained.
It is important to note that Mean deviation helps with a large
dataset with various values which is especially the case in the
stock market.
Going ahead, Variance is a related concept and is further
explained
 Variance is a dispersion measure which
suggests the average of differences from the
mean, in a similar manner as Mean
Deviation does, but here the deviations are
squared.
So,

Variance
Here, N = number of values in data set and
D0, D1, D2, D3 are the deviation of each
value in the data set from the mean
 In simple words, the
standard deviation is a
calculation of the spread out
Standard of numbers in a data set.
The
Deviation symbol (sigma)represents
Standard deviation and the
formula is:

Also,

is the formula of standard

deviation.
Visualization
Visualization helps the analysts to decide on the basis of
organized data distribution. There are four such types of
Visualization approach, which are:
• Histogram
• Bar Chart
• Pie Chart
• Line Chart
Probability Theory

• Now let us go back in time and recall the example of finding

probabilities of a dice roll. This is one finding that we all have
studied. Given the numbers on dice i.e. 1,2,3,4,5, and 6, the
probability of rolling a 1 is 1 out of 6 or ⅙. Such a probability is
known as discrete in which there are a fixed number of results.
• Now let us go back in time and recall the example of finding
probabilities of a dice roll. This is one finding that we all have
studied. Given the numbers on dice i.e. 1,2,3,4,5, and 6, the
probability of rolling a 1 is 1 out of 6 or ⅙. Such a probability is
known as discrete in which there are a fixed number of results.
• For discrete probabilities, there are certain cases which are so
extensively studied, that their probability distribution has
become standardised. Let’s take, for example, Bernoulli's
distribution, which takes into account the probability of getting
heads or tails when we toss a coin. We write its probability
function as px (1 – p)(1 – x). Here x is the outcome, which could
be written as heads = 0 and tails = 1.
• Now, let us look into the Monte Carlo Simulation in
understanding how it approaches the possibilities in the future,
taking a historical approach. It is said that the Monte Carlo
method is a stochastic one (in which there is sampling of
random inputs) to solve a statistical problem.
• Well, simply speaking, Monte Carlo simulation believes in
obtaining a distribution of results of any statistical problem or
data by sampling a large number of inputs over and over again.
Also, it says that this way we can outperform the market without
any risk.
• One example of Monte Carlo simulation can be rolling a dice
several million times to get the representative distribution of
results or possible outcomes. With so many possible outcomes,
it would be nearly impossible to go wrong with the prediction of
actual outcome in future. Ideally, these tests are to be run
efficiently and quickly which is what validates Monte Carlo
simulation.
• Although asset prices do not work by rolling a dice, they also
resemble a random walk. Let us learn about Random walk
now.
What is Random walk?
• Random walk suggests that the changes in stock prices have
the same distribution and are independent of each other.
Hence, based on the past trend of a stock price, future price
can not be predicted. Also, it believes that it is impossible to
outperform the market without bearing some amount of risk.
• Coming back to Monte Carlo simulation, it validates its own theory by
considering a wide range of possibilities and on the assumption that
it helps reduce uncertainty. Monte Carlo says that the problem is
when only one roll of dice or a probable outcome or a few more are
taken into consideration. Hence, the solution is to compare multiple
future possibilities and customize the model of assets and portfolios
accordingly.

• After the Monte Carlo simulation, it is also important to

understand Bayes’ theorem since it looks into the future
probabilities based on some relatable past occurrences and hence,
has usability.
• In simple words, Bayes’ theorem displays the possibility of the
occurrence of an event based on the past conditions that might have
led to a relatable event to take place.
• This can be applied in probability theory, wherein, based on the
past occurrences with regard to stock prices, the future ones
can be predicted.
Linear Algebra
• Matrices
• Vectors
• Programs
Linear Regression
• ML
Calculus
• Differential Calculus - It calculates the instantaneous change in
rates and the slopes of curves.
• Integral Calculus - This one calculates the quantities summed
up together.
Club with probability and graphics
Summary
• Descriptive Statistics
• Probability Theory
• Linear Algebra
• Linear Regression
• Calculus
Classification of Algorithmic Trading Strategies
A lot of algorithmic trading strategies that are being used today can be
classified broadly into the following two categories:
• ● Momentum based Strategies or Trend Following Strategies
• ● Mean reversion based strategies
Momentum Based Strategies
• Momentum trading involves trying to profit from the trends in asset
prices by taking a position in the direction of the trend. The rationale
behind such a strategy is that once the asset price gains momentum
in a particular direction, it keeps going in that direction. Traders spend
a lot of time and effort to determine the strength of these trends,
before they take a position to ride them. There are various technical
indicators that have been designed to gauge the strength and
direction of these trends using different approaches.
• Examples of such indicators include:
• ● Moving averages (both simple and exponential)
• ● Relative strength index (RSI)
• ● Moving average convergence divergence (MACD)
• ● Stochastic oscillator
Different automated trading strategies can be designed using these
indicators and the current price series of the asset.
Moving average crossover
Simple Moving Average
Simple Moving Averages are the averages of a series of numeric values.
They have a predefined length for the number of values to average at
each step called the ‘window size’. Given a series of numbers and a
fixed window size, the first element of the MA series is obtained by
taking the average of the first subset of the number series whose size
equals the window. The next value in MA series is obtained by
averaging the set of values when the window moves by one place i.e.
excluding one data point from the left and including one from the right
and so on.
• Consider the example mentioned below to understand the calculation
of simple moving averages. Let the average be calculated for five data
points (fixing the window size as 5). Suppose we want to find the MA
for the following price series: Daily price series for an asset: 7, 12, 2,
14, 15, 16, 11, 20, 7, 10, 23 Then, the first value of the MA series is
the average of first 5 data points= (7 + 12 + 2 + 14 + 15) / 5 = 10
Second value of the MA series is calculated as: (12 + 2 + 14 + 15 + 16)
/ 5 = 11.8 Similarly, the third value of the MA series is: (2 + 14 + 15 +
16 + 11) / 5 = 11.6 and so on.
The Simple Moving Average Crossover
Strategy
• When calculated for an asset price series, the MA series represents
the overall momentum or trend by ‘smoothing’ what can appear to
be ‘noisy’ or erratic price movements on short-term charts, into more
easily understood visual trend lines. Let us look at a simple example
of a moving average crossover strategy on the NIFTY index
• . The strategy involves moving average indicators of different
durations. An average of the shorter lookback window is called SMA
and the one with the longer look-back window is called LMA.
Popularly used SMA-LMA pairs include 20-40, 20-60 and 50-200. We
show below the SMA-LMA plot along with the adjusted close price for
Nifty
• Trading rules are simple. Buy the asset when the SMA crosses above LMA and sell
the asset when SMA crosses the LMA from above. The idea is to capture the trend
and profit from it as discussed above.
• In terms of the coding logic, we shall buy the asset when SMA(today) > LMA(today)
and SMA(yesterday)
• < LMA(yesterday). Similarly sell the asset when SMA(today) < LMA(today) and
SMA(yesterday) > LMA(yesterday).
• The parameters of this strategy are the SMA and LMA lookback windows that need
to be optimized. For example, 20-day SMA and 50 day LMA pair (20-50) may
perform better than a 15-50 pair. Decisions about strategy parameters can only be
taken after backtesting the strategy on past data and forward testing it on real time
data.
• Such strategies can be easily coded, visualized, backtested and executed using
programming languages such as Python.
Mean Reversion Based Strategies

Mean reversion trading is an approach which suggests

that prices, returns, or various economic indicators tend
to move to the historical average or mean over time. This
theory has led to many trading strategies which involve
the purchase or sale of a financial instrument whose
recent performance has greatly differed from their
historical average without any apparent reason.
Consider a hypothetical example of a commodity whose price
has stayed around USD 100 per tonne for the last ten years. Now,
if one day it’s price increases by USD 40 without any significant
news or factor behind this rise, then by the principle of mean
reversion, we can expect the price to fall in the coming days. In
such a case, the mean reversionist would sell the commodity, in
anticipation of the price falling in the following days. She would
make profits by buying back when the price has fallen back to its
mean of USD 100. The risk involved in this trade is that the price
may keep drifting away from the mean or stay away from the
mean longer than a trader can hold her position
Time Series
and
Stationarity
Time series is the data on a
particular variable (for e.g. an
asset prices or returns),
arranged in the order of time. A
mean-reverting time series has
been plotted below where the
horizontal black line represents
the mean and the blue curve is
the time series observations
which tends to revert back to
the mean.

•
•
• Loosely speaking, a time series is stationary if its mean
and variance are time invariant (constant over time). A
stationary time series will be mean reverting in nature,
i.e. it will tend to return to its mean and fluctuations
around the mean will have roughly equal amplitudes. A
stationary time series will also not drift too far away from
its mean
• A non-stationary time series, on the contrary, will have a time
varying variance or a time varying mean or both, and will often
not revert back to its mean.
• In the following diagram the blue line represents a non-
stationary time series, whereas the red line represents a
stationary time series.
• In the financial industry, traders take advantage of
stationary time series by placing orders when the
price of a security deviates considerably from its
historical mean, speculating the price to revert
back to its mean.
• One of the simplest mean reversion trading related trading strategies is
to find the average price over a specified period, followed by determining
a high-low range around the average value from where the price tends to
revert back to the mean. The trading signals will be generated when these
ranges are crossed - placing a sell order when the range is crossed on
the upper side and a buy order when the range is crossed on the lower
side. The trader takes contrarian positions, i.e. goes against the
movement of prices (or trend), expecting the price to revert back to the
mean.
•
Pairs Trading Strategy: Mean Reversion of Spreads

Pairs Trading is another strategy that relies on the

principle of mean reversion trading. Two assets are
said to be co-integrated if the difference (spread)
of their price series is stationary and hence mean
reverting. Assets whose prices move closely
together often exhibit co-integration, but not
always and we need to confirm this using
statistical tests
• For example, let us consider the stocks of two oil companies A
and B, which are fairly similar in terms of size, company
structures, markets and risk exposures. After quantitative
analysis, we have observed that the spread between the prices
of these stocks has remained fairly constant with a mean of
zero in the past five years, with occasional divergences. We
have performed statistical tests that have confirmed that the two
price series are cointegrated i.e. the spread is mean reverting.
• Now suppose suddenly the price of stock A increases
considerably without any apparent reason or market news,
whereas stock B continues to trade around its original price.
This increases the spread between the two securities,
significantly away from its mean value of zero to a value of
+2.5.
• According to our strategy we need to sell the spread. To do so,
we can then sell the overpriced asset (A) and buy the
underpriced asset (B) to enter into a pairs trade and wait for the
spread to come back to its historic mean value of 0.
• This strategy can easily be automated using a
programming language such as Python. We can write a
program to fetch price data for different stocks and
conduct co-integration checks using available libraries
and functions. Then, we can generate trading signals to
buy the underpriced security and sell the overpriced
security, when the spread breaches some threshold
values. We can also set appropriate stop-losses and
profit booking commands in place
Backtesting
• Backtesting is the process of validating our strategy by testing its
performance on historical data. It helps us to gauge the past performance
of the strategy. Backtesting also gives us the opportunity to optimize the
strategy parameters.
•
Through backtesting, we can measure strategy performance based on
certain metrics such as dollar PnL, percentage of profitable trades,
Sharpe ratio (a measure of risk-adjusted returns), maximum drawdown
(maximum fall in the value of the asset from a peak value), etc.
• Algorithmic traders spend most of their time researching and backtesting
their trading strategies using historical market data and other datasets as
required by the strategy
• One of the aspects to be considered in backtesting is the
‘backtesting window’ or how far back in the past should we go to
test our strategy. In the case of HFT, we generally don’t go back
to even the past 5 or 10 years of data because the market
microstructure changes much faster. For example, if you are
trying to backtest a simple strategy on data from 2007 when
algorithmic trading was not allowed in India, then the results
that we get will not be very useful. Typically, the backtesting
window for high frequency trading strategies is short compared
to the backtesting window for low frequency strategies that may
range from a few years to few decades.
• These days there are various platforms available which provide functionalities
to perform backtesting on historical data. The important points to consider
before selecting a backtesting platform are:
● knowing which asset classes and markets the platform supports,
•
● knowing about its sources of market data feeds and,
•
● figuring out which programming languages it supports.
•
• Backtesting allows us to test our strategies before actually implementing them
in the live market. However, the maxim ‘past performance does not necessarily
guarantee future returns’ should be kept in mind while backtesting.
Next
• More theoretical time series-ARIMA, GARCH
• Bayesian probability
• Backtesting
• Mean reversion
• Candle sticks
280421
Time Series
 Time Series

•A time series is a sequential set of data points, measured typically over

successive times.
•Time series analysis comprises methods for analyzing time series data in
to extract meaningful statistics and other characteristics of the data.
The impact of time series analysis on scientific applications can be partially
documented by producing an abbreviated listing of the diverse fields in
which important time series problems may arise. For example, many familiar
time series occur in the field of economics, where we are continually
exposed to daily stock market quotations or monthly unemployment figures.
Social scientists follow population series, such as birthrates or school
enrollments. An epidemiologist might be interested in the number of
influenza cases observed over some time period. In medicine, blood pressure
measurements traced over time could be useful for evaluating drugs used in
treating hypertension. Functional magnetic resonance imaging of brain-wave
time series patterns might be used to study how the brain reacts to certain
stimuli under various experimental conditions
• In our view, the first step in any time series investigation always involves
careful examination of the recorded data plotted over time. This scrutiny
often suggests the method of analysis as well as statistics that will be of use
in summarizing the information in the data. Before looking more closely at
the particular statistical methods, it is appropriate to mention that two
separate, but not necessarily mutually exclusive, approaches to time series
analysis exist, commonly identified as the time domain approach and the
frequency domain approach. The time domain approach views the
investigation of lagged relationships as most important (e.g., how does
what happened today affect what will happen tomorrow), whereas the
frequency domain approach views the investigation of cycles as most
important (e.g., what is the economic cycle through periods of expansion
and recession). We will explore both types of approaches in the following
sections.
 Categories and
Terminologies
• Time-domain vs. Frequency-domain
–Time-domain approach: how does what happened today
affect what will happen tomorrow?
These approaches view the investigation oflagged
relationships as most important, e.g. autocorrelation
analysis.
–Frequency-domain approach: what is the economiccycle
through periods of expansion and recession?
These approaches view the investigation ofcyclesas
most important, e.g. spectral analysis and wavelet
analysis.

5/77
 Categories and
Terminologies (cont.)
• univariate vs. multivariate
A time series containing records of a single variable is
termed as univariate, but if records of more than one
variable are considered then it is termed as multivariate.
• linear vs. non-linear
A time series model is said to be linear or non-linear
depending on whether the current value of the series
is a linear or non-linear function of past observations.
• discrete vs. continuous
In a continuous time series observations are measured at
every instance of time, whereas a discrete time series
contains observations measured at discrete points in time.
• This lecture will focus on univariate, linear, discrete
time series.
 Important Characteristics
Some important questions to first consider when first looking at a time series
are:
• Is there a trend, meaning that, on average, the measurements tend to
increase (or decrease) over time?
• Is there seasonality, meaning that there is a regularly repeating pattern of
highs and lows related to calendar time such as seasons, quarters, months,
days of the week, and so on?
• Are there outliers? In regression, outliers are far away from your line. With
time series data, your outliers are far away from your other data.
• Is there a long-run cycle or period unrelated to seasonality factors?
• Is there constant variance over time, or is the variance non-constant?
• Are there any abrupt changes to either the level of the series or the
variance?
 Components of aTime Series
• In general, a time series is affected by four components,
i.e. trend,seasonal,cyclical and irregular components.
– Trend
The general tendency of a time series to increase,
decrease or stagnate over a long period of time.

110
cents per pound

60 70 80 90

2005 2010 2015

The price of chicken: monthly whole bird spot price, Georgia docks, US
cents per pound, August 2001 to July 2016, with fitted linear trend line.
Components of a Time Series (cont.)
• In general, a time series is affected by four components,
i.e. trend,seasonal,cyclicalandirregularcomponents.
– Seasonal variation
This component explains fluctuations within a year during
the season, usually caused by climate and weather
conditions, customs, traditional habits, etc.

Quarterly Earnings per Share

15
9/77

10
5
0

1960 1965 1970 1975 1980

Johnson & Johnson quarterly earnings per share, 84 quarters, 1960-I to 1980-IV.
Components of a Time Series (cont.)
• In general, a time series is affected by four components,
i.e. trend,seasonal,cyclical and irregular components.
– Cyclical variation
This component describes the medium-term changes
caused by circumstances, which repeat in cycles. The
duration of a cycle extends over longer period of time.

130
10/77

110
90
70

1970 1972 1974 1976 1978 1980

Average weekly cardiovascular mortality in Los Angeles County. There are
508 six-day smoothed averages obtained by filtering daily values over the
10 year period 1970-1979.
 Components of a Time
Series (cont.)
• In general, a time series is affected by four components,
i.e. trend,seasonal,cyclical andirregula rcomponents.
– Irregular variation
Irregular or random variations in a time series are caused by
unpredictable influences, which are not regular and also do
11/77

not repeat in a particular pattern.

These variations are caused by incidences such as war,
strike,
earthquake, flood, revolution, etc.
There is no defined statistical technique for measuring
random fluctuations in a time series.
 Combination of Four
Components
• Considering the effects of these four components, two
different types of models are generally used for a time
series.

–Additive Model

12/77 Y(t) = T(t) + S(t) + C(t) + I(t)

Assumption: These four components are independent of

each other.

–Multiplicative Model

Y(t) = T(t) × S(t) × C(t) × I(t)

Assumption: These four components of a time series are

not necessarily independent and they can affect one
another.
Types of Models

There are two basic types of “time domain” models.

1.Models that relate the present value of a series to past values
and past prediction errors - these are called ARIMA models (for
Autoregressive Integrated Moving Average). We’ll spend
substantial time on these.
2.Ordinary regression models that use time indices as x-
variables. These can be helpful for an initial description of the
data and form the basis of several simple forecasting methods.
 Time Series Example:
White Noise
• White Noise
–A simple time series could be a collection
ofuncorrelated random variables, {wt}, with zero mean
µ = 0 andσw2finite
variance , denoted as wt ∼ wn(0, σ w2).
• Gaussian White Noise
–A particular useful white noise is Gaussian white noise,
14/77

wherein the wt areindependent normal random

and variance σw2mean
variables(with 0
), denoted as wt ∼ iid N (0, σ w2).
• White noise time series is of great interest because if
the stochastic behavior of all time series could be
explained in terms of the white noise model, then
classical statistical methods would suffice.
Time Series Example: Random Walk
• A random walk is the process by which randomly-
moving objects wander away from where they started.
• Consider a simple 1-D process:
–The value of the time series at time t is the value of the
series at time t − 1 plus a completely random movement
determined by wt . More generally, a constant drift factor δis
introduced. 15/77

Σt
Xt = δ+ X t− 1 + w t = δt + wi
i=1
−20 0 20 40 60 80

0 100 200 300 400 500

 Time Series Analysis
• The procedure of using known data values to fit a time
series with suitable model and estimating the
corresponding parameters. It comprises methods that
attempt to understand the nature of the time series and is
often useful for future forecasting and simulation.
• There are several ways to build time series forecasting
16/77

models,
but this lecture will focus on stochastic process.
–We assume a time series can be defined as a collection of
random variablesindexed according to the order they are
obtained in time, X1, X2, X3, ... t will typically be discrete
and vary over the integers t = 0, ±1, ±2, ...
–Note that the collection of random variables {Xt} is referred to
as astochastic process, while the observed values are
referred to as arealizationof the stochastic process.
 Measures of Dependence
• A complete description of a time series, observed as a
collection of n random variables at arbitrary time points
t1, t2, ...,tn, for any positive integer n, is provided by
the
joint distribution function, evaluated as the probability
that the values of the series are jointly less than the n
17/77

constants, c1, c2, ...,cn; i.e.,

Ft1,t2,...,tn (c1, c2, ...,cn) = Pr(Xt1 ≤ c1, Xt2 ≤ c2, ...,Xtn ≤ cn).

• Unfortunately, these multidimensional distribution

functions cannot usually be written easily.
• Therefore some informative descriptive measures
can be useful, such asmean functionand more.
 Measurement Functions
• Mean function
–The mean function is defined
as ∞
µt = µ Xt = E[X t] = ∫ xft (x)dx,
−∞

provided it exists, where E denotes the usual expected

18/77

value operator.
• Clearly for white noise series, µwt = E[wt ] = 0 for all t.
• For random walk with drift (δ ƒ= 0),
t
µXt = E[X t] = δt + Σ E[wi ] = δt
i=1
 Autocovariance for
•
Time Series
Lack of independence between adjacent values in time
series
Xs and Xt can be numericallyassessed.
• Autocovariance Function
–Assuming the variance of Xt is finite, the
autocovariance function is defined as the second
moment product
19/77

γ(s, t) = γX (s, t) = cov(Xs ,Xt ) = E[(Xs − µs)(Xt − µt)], for

all s and t.
–Note that γ(s, t) = γ(t, s) for all time points s andt.
• The autocovariance measures thelinear dependence
between two points on the same series observed at
different times.
–Very smooth series exhibit autocovariance functions that
stay large even when the t and s are far apart, whereas
choppy series tend to have autocovariance functions
 Autocorrelation for
Time Series
• Autocorrelation Function (ACF)
–The autocorrelation function is defined
as
γ(s,t)
ρ(s, t) = √
γ(s, s)γ(t, t)

–According to Cauchy-Schwarz inequality

20/77

2
|γ(s, t)| ≤ γ(s, s)γ(t, t),

it’s easy to show that −1 ≤ ρ(s, t) ≤ 1.

• ACF measures thelinear predictabilityofXt using only Xs .
–If we can predict Xt perfectly from Xs through a linear
relationship, then ACF will be either +1 or −1.
 Stationarity of
Stochastic Process
• Forecasting is difficult as time series is non-deterministic in nature,
i.e. we cannot predict with certainty what will occur in the future.
• But the problem could be a little bit easier if the time seriesis stationary:
you simply predict its statistical properties will be the same in the future as they have been
in the past! 21/77

–A stationary time series is one whose statistical properties such as mean, variance,
autocorrelation, etc. are all constant over time.
• Most statistical forecasting methods are based on the assumption that the time series
can be rendered approximately stationary after mathematical transformations.
 Which of these are
stationary?

22/77
 Strict
Stationarity
• There are two types of stationarity, i.e.strictly
stationaryand weakly stationary.
• Strict Stationarity
–The time series {Xt, t ∈Z} is said to be strictly stationary if
thejoint distributionof ( Xt1 ,Xt2 ,...,Xtk ) is thesameas that
of (Xt1+h, Xt2+h, ...,Xtk +h).
23/77

–In other words, strict stationarity means that the joint

distribution only depends on the “difference” h, not the
time (t1, t2, ...,tk ).
• However in most applications this stationary
condition is too strong.
 Weak Stationarity
• Weak Stationarity
–The time series {Xt, t ∈Z} is said to be weakly stationaryif
1 E [X t2]< ∞, ∀t ∈Z;
2 E [Xt ]= µ, ∀t ∈Z;
3 γX (s, t) = γX (s + h, t + h), ∀s, t, h ∈Z.
–In other words, a weakly stationary time series {Xt} must
24/77

have three features: finite variation, constant first moment,

and that the second moment γX (s, t) only depends on |t −
s| and not depends on s or t.
• Usually the term stationary means weakly stationary, and
when people want to emphasize a process is stationary in
the strict sense, they will usestrictly stationary.
 Remarks on
Stationarity
• Strict stationarity does not assume finite variance thus
strictly stationary does NOT necessarily imply weakly
stationary.
–Processes like i.i.d Cauchy is strictly stationary but not
weakly stationary.
• A nonlinear function of a strictly stationary time series is
25/77

still strictly stationary, but this is not true for weakly

stationary.
• Weak stationarity usually does not imply strict stationarity
as higher moments of the process may depend on time t.
• If time series {Xt} is Gaussian (i.e. the distribution functions
of {Xt} are all multivariate Gaussian), then weakly stationary
also implies strictly stationary. This is because a
multivariate Gaussian distribution is fully characterized by
its first two moments.
Autocorrelation for stationary Time Series
• Recall that the autocovariance γX (s, t) of stationary time
series depends on s and t only through |s − t|, thus we can
rewrite notation s = t + h, where h represents the time
shift.

γX (t + h, t) = cov(Xt+h, Xt) = cov(Xh, X0) = γ(h, 0) = γ(h)

26/77

• Autocovariance Function of Stationary Time Series

γ(h) = cov(Xt+h, Xt) = E[(Xt+h − µ)(Xt − µ)]

• Autocorrelation Function of Stationary Time Series

γ(t + h,t) γ(h)
ρ(h) = √ =
γ (t + h, t + h)γ(t, t) γ(0)
 Partial Autocorrelation
• Another important measure is called partial
autocorrelation, which is the correlation between Xs and
Xt with the linear effect of “everything in the middle”
removed.
• Partial Autocorrelation Function (PACF)
–For a stationary process Xt , the PACF (denoted as φhh ), for
h = 1, 2, ... is defined as
27/77

φ11 = corr(Xt+1, Xt ) = ρ1
φhh = corr(Xt+h − Xˆt+h, Xt −Xˆt), h ≥2

where Xˆt+hand Xˆtis defined as:

Xˆt+h = β1Xt+h−1 + β2Xt+h−2 + ··· + βh−1Xt+1

Xˆt= β1Xt+1 + β2Xt+2 + ··· + β h−1 X t+h−1
–If Xt is Gaussian, then φhh is actually conditional correlation
φhh = corr(Xt, Xt+h|Xt+1, Xt+2, ...,X t+h−1 )
 ARIMA Models
• ARIMA is an acronym that stands for Auto-
Regressive Integrated Moving Average.
Specifically,
– AR Autoregression. A model that uses the dependent
relationship between an observation and some
number of lagged observations.
– I Integrated. The use ofdifferencingof raw observations
28/77

in order to make the time series stationary.

– MA Moving Average. A model that uses the dependency
between an observation and aresidual errorfrom a
moving average model applied to lagged
observations.
• Each of these components are explicitly specified in the
model as a parameter.
• Note that AR“stationarize”
procedureto and MA are twotimewidely
seriesusedlinear
if
modelsthat
needed. work on stationary time series, and I is
apreprocessing
 Notations
• A standard notation is used of ARIMA(p, d, q) where
the parameters are substituted with integer values to
quickly indicate the specific ARIMA model being
used.
– p The number of lag observations included in the model,
also called thelag order.
– d The number of times that the raw observations
are differenced, also called thedegree of
differencing.
– q The size of the moving average window, also called
theorder
of moving average.
• A value of 0 can be used for a parameter, which indicates
to not use that element of the model.
• In other words, ARIMA model can be configured to
perform the function of an ARMA model, and even a 26/77
 Autoregressive Models
• Intuition
–Autoregressive models are based on the idea that current
value of the series, Xt , can be explained as alinear
combinationof p past values, X t−1 , X t−2 , ..., X t−p ,
errorinwith
together the same series.
arandom
• Definition
–An autoregressive model of order p, abbreviated AR(p), is
30/77

of the form
Σp
Xt = φ 1X t− 1 + φ 2X t− 2 + ··· + φ X
p t− p +w t = φX
i t− i +w t
i=1

where Xt is stationary, w t∼ wn(0, σ ),w2 and φ, φ

1 , . 2. . , φ p
(φp ƒ= 0) are model parameters. The hyperparameter p
represents the length of the “direct look back” in the
series.
 Backshift Operator
• Before we dive deeper into the AR process, we need some
new notations to simplify the representations.
• Backshift Operator
–The backshift operator is defined as

BXt =X t−1 .
31/77

It can be extended, B2Xt = B(BXt ) = B(X t−1 ) = X t−2 , and

so on. Thus,
B k Xt = X t−k
• We can also define an inverse operator (forward-
shift operator) by enforcing B −1 B = 1, such that

Xt = B −1 BX t =B −1 X t−1 .
Autoregressive Operator of AR Process
• Recall the definition for AR(p) process:
Xt = φ1Xt−1 + φ2Xt−2 + ··· + φpX t−p + wt By

using the backshift operator we can rewrite it as:

Xt −φ1Xt−1 −φ2Xt−2 − ··· −φpX t−p = wt

(1 − φ1B − φ2B2 −··· − φpB p )Xt = wt
32/77

• Theautoregressive operatoris defined as:

Σp
φ(B) = 1 − φ B1 − φ B2 − 2· · · − φ B = p
p 1− φjB j,
j=1

then the AR(p) can be rewritten more concisely

as:
φ(B)Xt =wt
 AR Example: AR(0)
and AR(1)
• The simplest AR process is AR(0), which has no dependence between the terms.
In fact, AR(0) is essentiallywhite noise.
• AR(1) can be given by Xt = φ1Xt−1 + wt.
• –Only the previous term in the process and the noise term
33/77

contribute to the output.

• –If |φ1| is close to 0, then the process still looks like whitenoise.
–If φ1< 0, Xt tends to oscillate between positive and negative values.
• –If φ1= 1 then the process is equivalent to random walk, which
is not stationary as the variance is dependent on t (and infinite).
 AR Examples: AR(1) Process
• Simulated AR(1) Process Xt = 0.9Xt−1 +wt :

34/77

• Mean E[Xt] = 0
σw2
• Varianc Var(X t) =
e (1 − φ)21
 AR Examples: AR(1)
Process
• Autocorrelation Function (ACF)
ρh =φh 1

35/77
 AR Examples: AR(1)
Process
• Partial Autocorrelation Function (PACF)

φ11 = ρ1 = φ1 φhh = 0, ∀h ≥ 2

36/77
 AR Examples: AR(1) Process
• Simulated AR(1) Process Xt = −0.9Xt−1 +wt :

37/77

• Mean E[Xt] = 0
σw2
• Varianc Var(X t) =
e (1 − φ)21
 AR Examples: AR(1)
Process
• Autocorrelation Function (ACF)
ρh =φh 1

38/77
 AR Examples: AR(1)
Process
• Partial Autocorrelation Function (PACF)

φ11 = ρ1 = φ1 φhh = 0, ∀h ≥ 2

39/77
 Stationarity of AR(1)
• We can iteratively expand AR(1) representation
as:

Xt ==φ Xt−1X+ wt +w
φ1(φ + w t= φ X1 + φ 1w t− 1 +w t
1 1 t− 2 t− 1 ) t− 2
2
..
40/77 Σ
k− 1
j
= φX
k
1 t− k + φw
1 t− j
j=0

• Note that if |φ1| < 1 and supt Var(Xt ) < ∞, we have:

Σ∞ j
Xt = φw
1 t− j
j=0

This representation is called the stationary

solution.
 AR Problem: Explosive
AR Process
• We’ve seen AR(1): Xt = φ1Xt−1 + wt while |φ1| ≤ 1.
• What if |φ1| > 1? Intuitively the time series will “explode”.
• However, technically it still can bestationary, because
we expand the representation differently and get:
41/77
Σ∞
Xt = − φ−1 j wt+j
j=1

• But clearly this is not useful because we need the

future (wt+j ) to predict now (Xt).
• We use the concept of causality to describe time series
that is not only stationary but also NOT future-dependent.
 General AR(p) Process

• An important property of AR(p) models in general is

–When h > p, theoretical partial autocorrelation function is0:

φhh = corr(Xt+h − Xˆt+h, Xt − Xˆt) = corr(wt+h,Xt − Xˆt) = 0.

42/77

–When h ≤ p, φpp isnot zeroand φ11, φ22, . . . , φh−1,h−1 are

not necessarily zero.
• In fact,identification of an AR model is often best done
with the PACF.
 AR Models:
Parameters Estimation
• Note that p is like a hyperparameter for the AR(p)
process, thus fitting an AR(p) model presumes p is
known and only focusing on estimatingcoefficients, i.e.
φ1,φ2,...,φp.
• There are many feasible approaches:
43/77

• Method of momentsestimator (e.g. Yule-Walkerestimator)

• Maximum Likelihood Estimation (MLE)estimator
• Ordinary Least Squares (OLS)estimator
• If the observed series is short or the process is far
from stationary, then substantial differences in the
parameter estimations from various approaches are
expected.
 Moving Average
Models (MA)
• The name might be misleading, butmoving average
models should not be confused with themoving
average smoothing.
• Motivation
–Recall that in AR models, current observation Xt is
44/77

regressed using the previous observations X t−1, Xt−2, . . . ,

X t−p , plus an error term wt at current time point.
–One problem of AR model is the ignorance of correlated
noise structures (which is unobservable) in the time
series.
–In other words, the imperfectly predictable terms in current
time, wt , and previous steps,wt−1, w t−2 , ...,w t−q ,
arealso informativefor predicting observations.
 Moving Average
Models (MA)
• Definition
–A moving average model of order q, or MA(q), is defined to
be
Σq
Xt = w t +θ w1 t − 1 + θ 2w t − 2 + ··· + θ w
q t−q = w t + θw
j t−j
j=1
45/77

where wt ∼ wn(0, σw2), and θ ,1θ, .2 . . , θ(θq = 0)are

qƒ
parameters.
–Although it looks like a regression model, the difference is
that the wt is not observable.
• Contrary to AR model, finite MA model isalways
stationary, because the observation is just a weighted
moving average over past forecast errors.
 Moving Average
Operator
• Moving Average Operator
–Equivalent to autoregressive operator, we define
moving average operator as:

θ(B) = 1 + θ1B + θ2B2 + ··· + θqBq,

where B stands for backshift operator, thus B(wt ) = w t−1 .

46/77

• Therefore the moving average model can be rewritten

as:

Xt = wt + θ1wt−1 + θ2wt−2 + ··· + θqwt−q

Xt = (1 + θ1B + θ2B2 + ··· + θqB q )wt
Xt =θ(B)wt
 MA Examples:
•
MA(1) Process
Simulated MA(1) Process X = w + 0.8 ×w
t t t−1 :

47/77

• Mean E[Xt] = 0
• Variance Var(Xt) = σ2 (1 + θ2)
w 1
 MA Examples: MA(1)
Process
θ1
• Autocorrelation Function (ACF)
ρ1 = ρh= 0, ∀h ≥ 2
1+ θ 1
2

48/77
MA Examples: MA(1) Process
• Partial Autocorrelation Function (PACF)
h
(−θ )1 (1 − θ) 2
1,
φhh =− 2(h+1)
h ≥1
1− θ 1

49/77
MA Examples: MA(2) Process
• Simulated MA(2) Process Xt = wt + 0.5×w t−1 + 0.3×w t−2 :

50/77

• Mean E[Xt] = 0
• Variance Var(Xt) = σ2 (1 + θ2 + θ2)
w 1 2
MA Examples: MA(2) Process
• Autocorrelation Function
(ACF) θ1 +θ1θ2 θ2
ρ1 = ρ2 = ρh= 0, ∀h ≥ 3
1 + θ 1+ θ 2
2 2 1 + θ21+ θ2 2

51/77
 General MA(q) Process

• An important property of MA(q) models in general is

that there arenonzero autocorrelationsfor the first q
lags, and ρh = 0 for all lags h >q.
• In other words, ACF provides a considerable amount of
52/77

information about the order of the dependence q for

MA(q) process.
• Identification of an MA model is often best done with
the ACF rather than the PACF.
MA Problem: Non-unique MA Process
• Consider the following two MA(1) models:
– Xt = wt + 0.2wt−1, wt ∼ iid N (0, 25),
– Yt = vt + 5v t−1 , vt ∼ iid N (0, 1)
5
• Note that both of them have Var(Xt ) = 26 andρ1 = .
2
• In fact, these two MA(1) processes are essentiallythe
53/77
6
same. However, since we can only observe Xt (Yt ) but
not noise terms wt (vt ), we cannot distinguish them.
• Conventionally, we define the concept invertibility and
always
choose the invertible representation from multiple
alternatives.
–Simply speaking, for MA(1) models the invertibility condition
is |θ1| < 1.
 Comparisons between AR and MA
• Recall that we have seen for AR(1) process, if |φ1| < 1
and supt Var(Xt )< ∞,

Σ∞ j
Xt = φw
1 t− j
j=0
54/77

• In fact, allcausal AR(p) processes can be represented as

MA(∞); In other words, infinite moving average
processes are finite autoregressive processes.
• Allinvertible MA(q) processes can be represented as
AR(∞).
i.e. finite moving average processes are
infinite autoregressive processes.
MA Models: Parameters Estimation
• A well-known fact is that parameter estimation for MA
model is more difficult than AR model.
–One reason is that the lagged error terms are not observable.
• We can still usemethod of momentsestimators for
MA process, but we won’t get the optimal
55/77

estimators with Yule-Walker equations.

• In fact, since MA process is nonlinear in the parameters,
we need iterative non-linear fitting instead of linear least
squares.
• From a practical point of view,modern scientific
computing software packages will handle most of the
details after given the correct configurations.
 ARMA Models
• Autoregressive and moving average models can be
combined together to form ARMA models.
• Definition
–A time series {xt;t = 0, ±1, ±2, ...} is ARMA(p, q) ifit is
stationaryand
56/77
Σp Σq
Xt = w t+ φX
i t−i + θw
j t−j ,
i=1 j=1

where φp=ƒ 0, θ q ƒ= 0, and σw2 > 0, wt ∼ wn(0, σ w2).

–With the help of AR operator and MA operator we
defined before, the model can be rewritten more
concisely as:
φ(B)Xt = θ(B)wt
 ARMA Problems:
Redundant Parameters
• You may have observed that if we multiply a same factor
on both sides of the equation, it still holds.

η(B)φ(B)Xt =η(B)θ(B)wt

• For example, consider a white noise process Xt = wt and

η(B) = (1− 0.5B):
57/77

(1 − 0.5B)Xt = (1− 0.5B)wt

Xt = 0.5Xt−1 − 0.5wt−1 + wt
• Now it looks exactly like a ARMA(1, 1)process!
• If we were unaware of parameter redundancy, we might
claim the data are correlated when in fact they are not.
 Choosing Model
Specification
• Recall we have discussed that ACF and PACF can be used
for determining ARIMA model hyperparamters p and q.

AR(p) MA(q) ARMA(p, q)

Cuts off
ACF Tails off Tails off
58/77
after lag q
Cuts off
PACF Tails off Tails off
after lag p

• Other criterions can be used for choosing q and q too, such

as AIC (Akaike Information Criterion), AICc (corrected AIC)
and BIC (Bayesian Information Criterion).
• Note that the selection for p and q is not unique.
“Stationarize” Nonstationary Time Series
• One limitation of ARMA models is thestationaritycondition.
• In many situations, time series can be thought of as
being composed of two components,a non-stationary
trend series anda zero-mean stationary series, i.e. Xt =
µt + Yt .
• Strategies
–Detrending: Subtracting with an estimate for trend and
59/77

deal with residuals.

Yˆt= Xt − µˆt
–Differencing: Recall that random walk with drift is capable
of representing trend, thus we can model trend as a
stochastic component as well.
µt = δ+ µ t−1 + wt
∇ X t = Xt − X t−1 = δ+ wt + (Yt − Y t−1 ) = δ+ wt +∇ Y t
∇ is defined as the first difference and it can be extended
to higher orders.
 Differencing
• One advantage of differencing over detrending for
trend removal is that no parameter estimation is
required.
• In fact, differencing operation can be repeated.
–The first difference eliminates a linear trend.
–A second difference, i.e. the difference of first difference,
can eliminate a quadratic trend.
60/77

• Recall the backshift operator Xt =BX t−1 :

∇ 2X∇ X t = Xt − X t−1 = Xt − BXt = (1 −B)Xt

t = ∇ ( ∇ X t ) = ∇ (X t − X t−1 )
= (Xt − X t−1 )− (X t−1 − X t−2 )
= Xt − 2X t−1 + X t−2 = Xt − 2BXt +B2Xt
= (1 − 2B + B2)Xt = (1 −B)2Xt
 Detrending vs. Differencing
original


  
 




 
 





chicken






60 70 80 90 110
   
  
    
    
 
 
    
   
 

  
  
  

 
 
    
   
 
   
         


2005 2010 2015

Time

detrended

 

61/77
 
   
 
   

   
 

 
  

resid(fit)

   


    
 
  
   
         
 
    
    
−5 0 5 10

 
    
 
     
 
   
  
 
 
  


  
  
   
 

  
 

      

 


2005 2010 2015

Time

first difference

 


  
 
 
 
 
  
diff(chicken)

  
  
  
  
 
     
    
     
   
    
   
 
        
     
−2 −1 0 1 2 3

       
     



    
      
    
 
    
   
  
  
   
   
  
  
 
  
 
 
 


  
  


2005 2010 2015

Time
 From ARMA to ARIMA
• Order of Differencing
–Differences of order d are defined as
d d
∇ = (1 − B) ,

where (1 − B)d can be expanded algebraically for

higher integer values of d.
• Definition 62/77

–A process Xt is said to be ARIMA(p, d, q)if

d d
∇ Xt = (1 − B) Xt

is ARMA(p, q).
–In general, ARIMA(p, d, q) model can be writtenas:

φ(B)(1 − B)d Xt =θ(B)wt

.
 Box-Jenkins Methods
• As we have seen ARIMA models have numerous
parameters and hyper parameters, Box and Jenkins
suggests an iterative three-stage approach to estimate
an ARIMA model.
• Procedures
1 Model identification: Checking stationarity and
seasonality, performing differencing if necessary,
63/77

choosing model specification ARIMA(p, d, q).

2 Parameter estimation: Computing coefficients that best fit
the selected ARIMA model using maximum likelihood
estimation or non-linear least-squares estimation.
3 Model checking: Testing whether the obtained model
conforms
to the specifications of a stationary univariate process (i.e.
the residuals should be independent of each other and
have constant mean and variance). If failed go back to step
1.
• Let’s go through a concrete example together for
 Air Passenger
Data

64/77

Monthly totals of a US airline passengers, from 1949 to 1960

 Model Identification
• As with any data analysis, we should construct a time plot
of the data, and inspect the graph for any anomalies.
• The most important thing in this phase is to determine if
the time series isstationaryand if there is anysignificant
seasonalitythat needs to be handled.
• Test Stationarity
65/77

–Recall the definition, if the mean or variance changes over

time then it’s non-stationary, thus an intuitive way is to
plotrolling statistics.
–We can also make anautocorrelation plot, as non-
stationary time series often shows very slow decay.
–A well-established statistical test calledaugmented
Dickey-Fuller Testcan help. The null hypothesis is the
time series is non-stationary.
 Stationarity Test: Rolling
Statistics

66/77

Rolling statistics with sliding window of 12 months

 Stationarity Test: ACFPlot

67/77

Autocorrelation with varying lags

 Stationarity Test: ADF Test
• Results of Augmented Dickey-Fuller
Test Item Value
Test Statistic 0.815369
p-value 0.991880
#Lags Used 13.000000
Number of Observations Used 130.000000
Critical Value (1%)
68/77
-3.481682
Critical Value (5%) -2.884042
Critical Value (10%) -2.578770

• The test statistic is a negative number.

• The more negative it is, the stronger the rejection of the
null hypothesis.
 Stationarize Time Series

• As all previous methods show that the initial time series

is non-stationary, it’s necessary to
performtransformationsto make it stationary for ARMA
modeling.
–Detrending
–Differencing
69/77

–Transformation: Applying arithmetic operations like log,

square root, cube root, etc. to stationarize a timeseries.
–Aggregation: Taking average over a longer time period,like
weekly/monthly.
–Smoothing: Removing rolling average from original time
series.
–Decomposition: Modeling trend and seasonality explicitly
and removing them from the time series.
Stationarized Time Series: ACF
Plot

70/77

First order differencing over logarithm of passengers

 Stationarized Time
Series: ADF Test
• Results of Augmented Dickey-Fuller
Test Item Value
Test Statistic -2.717131
p-value 0.071121
#Lags Used 14.000000
Number of Observations Used 128.000000
71/77

Critical Value (1%) -3.482501

Critical Value (5%) -2.884398
Critical Value (10%) -2.578960

• From the ACF plot, we can see that the mean and
std variations have much smaller variations with
time.
• Also, the ADF test statistic is less than the 10% critical
value, indicating the time series is stationary with 90%
confidence.
 Choosing Model
Specification

72/77

• Firstly we notice an obvious peak at h = 12, because

for simplicity we didn’t model the cyclical effect.
• It seems p = 2, q = 2 is a reasonable choice. Let’s see
three models, AR(2), MA(2) and ARMA(2, 2).
 AR(2): Predicted
on Residuals

73/77

• RSS is a measure of the discrepancy between the data

andthe estimation model.
–A small RSS indicates a tight fit of the model to the data.
 MA(2): Predicted on
Residuals

74/77
ARMA(2, 2): Predicted on Residuals

75/77

• Here we can see that the AR(2) and MA(2) models

have almost the same RSS but combined is
significantly better.
 Forecasting
• The last step is to reverse the transformations we’ve done
to get the prediction on original scale.

76/77
SARIMA: Seasonal ARIMA Models
• One problem in the previous model is the lack of
seasonality, which can be addressed in a generalized
version of ARIMA model calledseasonal ARIMA.
• Definition
–A seasonal ARIMA model is formed by including
additional seasonal terms in the ARIMA models,
denoted as 77/77

ARIMA (p, d, q)( P, D, Q)m,

i.e.
φ(Bm) φ(B)(1 − B m )D (1 − B)d Xt = θ(Bm) θ(B) wt

where m represents the number of observations per

year.
–The seasonal part of the model consists of terms that are
similar to the non-seasonal components, but involve
ARCH and GARCH
• As an example of financial time series data, Figure shows the daily
returns (or percent change) of the Dow Jones Industrial Average (DJIA)
from April 20, 2006 to April 20, 2016. It is easy to spot the financial
crisis of 2008 in the figure. The data shown in Figure are typical of
return data. The mean of the series appears to be stable with an
average return of approximately zero, however, highly volatile
(variable) periods tend to be clustered together. A problem in the
analysis of these type of financial data is to forecast the volatility of
future returns. Models such as ARCH and GARCH models (Engle, 1982;
Bollerslev, 1986) and stochastic volatility models (Harvey, Ruiz and
Shephard, 1994) have been developed to handle these problems.
DJIA returns
-.05 to 0.10

April21 2006 to March 31 2016

• The data were obtained using the Technical Trading Rules (TTR)
package to download the data from YahooTM and then plot it. We
then used the fact that if xt is the actual value of the DJIA and rt =
(xt−xt−1)/xt−1 is the return, then 1 + rt = xt/xt−1 and log(1 + rt) =
log(xt/xt−1) = log(xt) − log(xt−1) ≈ rt . 1.2 The data set is also available
in astsa, but xts must be loaded. # library(TTR) # djia =
getYahooData("^DJI", start=20060420, end=20160420, freq="daily")
library(xts) djiar = diff(log(djia$Close))[-1] # approximate returns
plot(djiar, main="DJIA Returns", type="n"lines(djiar)
 Off-the-shelf Packages
• Rhas arima function in standard packagestats.
• Mathematicahas a complete library of time series
functions including ARMA.
• MATLABincludes functions such as arima.
• Pythonhas a statsmodels module provides time series
81/77
analysis including ARIMA. Another Python module
called pandas provides dedicated class for time series
objects.
• STATAincludes the function arima for estimatingARIMA
models.
• SAShas an econometric package, ETS, for ARIMA
models.
• GNU Octavecan estimate AR models using functions
from extra packages.
08-05-2021
GARCH/ARCH
Modeling Unequal Variability
• Equal Variability: Homoscedasticity

• Unequal Variability: Heteroscedasticity

• Means any variability (around the mean) that is not
homoscedasticity
• Models must be developed for specific cases
The assumption of homoscedasticity (meaning “same variance”)
is central to linear regression models. Homoscedasticity
describes a situation in which the error term (that is, the “noise”
or random disturbance in the relationship between the
independent variables and the dependent variable) is the same
across all values of the independent
variables. Heteroscedasticity (the violation of homoscedasticity)
is present when the size of the error term differs across values of
an independent variable. The impact of violating the assumption
of homoscedasticity is a matter of degree, increasing as
heteroscedasticity increases.
A simple bivariate example can help to illustrate
heteroscedasticity: Imagine we have data on family income and
spending on luxury items. Using bivariate regression, we use
family income to predict luxury spending. As expected, there is a
strong, positive association between income and
spending. Upon examining the residuals we detect a problem –
the residuals are very small for low values of family income
(almost all families with low incomes don’t spend much on luxury
items) while there is great variation in the size of the residuals for
wealthier families (some families spend a great deal on luxury
items while some are more moderate in their luxury
spending). This situation represents heteroscedasticity because
the size of the error varies across values of the independent
variable. Examining a scatterplot of the residuals against the
predicted values of the dependent variable would show a classic
cone-shaped pattern of heteroscedasticity.
The problem that heteroscedasticity presents for regression
models is simple. Recall that ordinary least squares regression
seeks to minimize residuals and in turn produce the smallest
possible standard errors. By definition, OLS regression gives
equal weight to all observations, but when heteroscedasticity is
present, the cases with larger disturbances have more “pull” than
other observations. In this case, weighted least squares
regression would be more appropriate, as it down-weights those
observations with larger disturbances.
• A more serious problem associated with heteroscedasticity is the fact
that the standard errors are biased. Because the standard error is
central to conducting significance tests and calculating confidence
intervals, biased standard errors lead to incorrect conclusions about
the significance of the regression coefficients. Many statistical
programs provide an option of robust standard errors to correct this
bias. Weighted least squares regression also addresses this
concern but requires a number of additional assumptions. Another
approach for dealing with heteroscedasticity is to transform the
dependent variable using one of the variance stabilizing
transformations. A logarithmic transformation can be applied to
highly skewed variables, while count variables can be transformed
using a square root transformation. Overall however, the violation of
the homoscedasticity assumption must be quite severe in order to
present a major problem given the robust nature of OLS regression.
What These Acronym Mean?
• ARCH
• Autoregressive Conditional Heteroscedasticity

• GARCH
• Generalized ARCH
Volatility = Standard Deviation

Unconditional- Time varying volatility or

volatility based on Historical data is not
possible.
Provides positive coefficients
Forecasting is not possible
Generalization is not possible
ARCH
Auto regressive Conditional heteroskedasticity- this
observe the heteroskedasticity over period is auto-
correlated.
An auto correlated time series is predictable,
Serial correlation/lagged correlation/autocorelation
 Why to choose an ARCH model?

The Problem of Heteroscedasticity

Define a linear regression model:
Yt  0  1 X t   t

One of the basic assumptions of OLS method is:

E ( t )  
2 2

Implies that the conditional variance of Yi is a constant or the errors (i) are
E ( t )   t
homocedastic.
2 On the
2 contrary, if

then the variance is not a constant or the errors are Heteroscedastic. This is a
common problem with cross section data.
If so, estimates (0 and 1) are unbiased, but conventional estimates of
standard errors and confidence intervals are very narrow, giving false sense of
precision. For instance, the usual variance formula is:

var(1 )   2  xt2
whereas the true variance with heteroskedastic errors is:
*
var(1 )   x  2
t t
2
 x 
2 2
t

 *
var( 
Obviously,
1 )  var( 1 )

Thus, application of OLS method provides biased variance estimate; hence,

inference based on OLS estimates will be misleading.
The Persistence of volatility
High frequency financial time series data exhibit volatility clustering, which
shows that some periods are highly volatile with some degree of
autocorrelation.

Forecasting Mean and Variance

0.1Nifty return (2/01/1996 - 20/12/2002)

0.05
0
-0.05
-0.1

1033
1119
1205
1291
1377
1463
1549
1635
1721
87
173
259
345
431
517
603
689
775
861
947
1

It is easy to imagine a situation in which prediction of volatility is highly

useful. For instance, asset holders want to forecast both mean and variance of
the return. In this context, ARCH models are highly useful.
ARCH Process
Thus far, we have assumed that the variance is a constant. However, time series data
with volatility clustering indicate that the variance is not a constant; hence, it is
essential to model the variance as an AR process. Suppose we estimate the model:

Yt  X t   t
2 2 2
    1 t 1  t 2
where t ist a white0 noise process. The2 conditional
t
variance for t is:

2
E t  0  1 t21   2 t22
The above equation is called an ARCH (2) process.

The most convenient way is to model both mean and variance simultaneously using
maximum likelihood techniques. Engle (1982) provides a multiplicative
conditionally heteroscedastic model:

where 
tt isawhite
t  0noise
  1process
 t 1 with unit variance; 0 > 0 and 0 < 1 < 1.
2
Drawbacks of ARCH P process
Consumes several degrees of freedom
Difficult to interpret if some coefficients are
negative
Does not estimate mean and variance functions
simultaneously
Thus an ARCH model higher than ARCH(3) is
better estimated by GARCH Model
The GARCH Model

Bollerslev (1986) defines conditional variance as an ARMA process, which is

called gerneralised ARCH model or GARCH model. The GARCH (1,1) model
is:
rt     t  t2
 t2   0   1 t21   2 t21

where t-1 and t-1 are ARCH and GARCH terms respectively.

Eviews uses the following maximum likelihood method to estimate these

parameters simultaneously:
1 1 1
lt   log( 2 )  log(  t2 )  ( yt  xt  ) 2 /  t2
'

2 2 2
The ARCH-M Model

In some of the applications, the dependent variable in the mean equation

may be determined by its conditional variance. For instance, the expected
return on an asset is related to expected asset risk. If so, the mean equation
can be defined as:

rt     t2   t  t2
 t2   0   1 t21   2 t21

Such models are called ARCH-in-Mean or ARCH-M Models due to

(Engle, Lillien, Robins, 1987).
The Asymmetric ARCH Models

Often, we notice that the downward movements in the market are highly
volatile than upward movement of the same magnitude. In such case, symmetric
ARCH model undermines the true variance process. Engle and Ng (1993)
provide a news impact curve with asymmetric response to good and bad news.

Volatility

Bad news Good news

Eviews supports estimating two types of asymmetric ARCH models – TARCH

and EGARCH.
The TARCH Model

The T ARCH or Threshold ARCH due to Zakoian (1990) can be defined as:

rt     t  t2
 t2   0   1 t21   2 t21   t21d t 1
where d t-1  1 if ε t  0 and 0 otherwise
ε t  0  good news, ε t  0  bad news
good news has an impact of 1 t2-1 while bad news has an
impact of (1   ) t2-1
If   0, there is leverage effect.
if   0, the news impact is asymmetric .
The EGARCH Model

The EGARCH or exponential GARCH model proposed by Nelson (1991)

takes the form:

 t 1  t 1
log    0   1 log 
2 2
t 1  3 
 t 1  t 1
t

This implies that the leverage effect is exponential than quadratic.

If  t-1  t-1 0 the impact is ( 3   )
if  t-1  t-1 0 the impact is (- 3   )
If  0, there is leverage effect.
if   0, the impact is asymmetric
Model parameters

The standard ARIMA models expect as input parameters 3

arguments i.e. p,d,q.
• p is the number of lag observations.
• d is the degree of differencing.
• q is the size/width of the moving average window.
•
Getting the stock price history data

• Thanks to Yahoo finance we can get the data for free. Use the
following link to get the stock price history
of TESLA: https://finance.yahoo.com/quote/TSLA/history?perio
d1=1436486400&period2=1594339200&interval=1d&filter=histo
ry&frequency=1d
• Dowload and save ,csv
• 2015-2020
• Modules needed: Numpy, Pandas, Statsmodels, Scikit-Learn
• Our imports
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from pandas.plotting import lag_plot
from pandas import datetime
from statsmodels.tsa.arima_model import ARIMA
from sklearn.metrics import mean_squared_error
Now let’s load the TESLA stock history data:
df = pd.read_csv("TSLA.csv")
df.head(5)
• Now let’s load the TESLA stock history data:
df = pd.read_csv("TSLA.csv")
df.head(5)
• Target variable is the close price
Before building the ARIMA model, let’s see if there is some cross-
correlation in out data.
.
plt.figure()
lag_plot(df['Open'], lag=3)
plt.title('TESLA Stock - Autocorrelation plot with lag = 3')
plt.show()
We can now confirm that ARIMA is going to be a good model to
be applied to this type of data (there is auto-correlation in the
data).
Stock price over time
.

plt.plot(df["Date"], df["Close"])
plt.xticks(np.arange(0,1259, 200), df['Date'][0:1259:200])
plt.title("TESLA stock price over time")
plt.xlabel("time")
plt.ylabel("price")
plt.show()
• Build the predictive ARIMA model
Next, let’s divide the data into a training (70 % ) and test (30%)
set. For this tutorial we select the following ARIMA parameters:
p=4, d=1 and q=0.
train_data, test_data = df[0:int(len(df)*0.7)], df[int(len(df)*0.7):]training_data = train_data['Close'].values
test_data = test_data['Close'].valueshistory = [x for x in training_data]
model_predictions = []
N_test_observations = len(test_data)for time_point in range(N_test_observations):
model = ARIMA(history, order=(4,1,0))
model_fit = model.fit(disp=0)
output = model_fit.forecast()
yhat = output[0]
model_predictions.append(yhat)
true_test_value = test_data[time_point]
history.append(true_test_value)MSE_error = mean_squared_error(test_data, model_predictions)
print('Testing Mean Squared Error is {}'.format(MSE_error))
Summary of the code
• We split the training dataset into train and test sets and we use
the train set to fit the model, and generate a prediction for each
element on the test set.
• A rolling forecasting procedure is required given the
dependence on observations in prior time steps for
differencing and the AR model. To this end, we re-create
the ARIMA model after each new observation is received.
• Finally, we manually keep track of all observations in a list
called history that is seeded with the training data and to which
new observations are appended at each iteration.
• Testing Mean Squared Error is 741.0594879572484
• The MSE of the test set is quite large denoting that the precise
prediction is a hard problem. However, this is the average
squared value across all the test set predictions.
Let’s visualize the predictions to understand the performance
of the model more.
test_set_range = df[int(len(df)*0.7):].indexplt.plot(test_set_range,
model_predictions, color='blue', marker='o’,
linestyle='dashed',label='Predicted Price')plt.plot(test_set_range, test_data,
color='red', label='Actual Price')plt.title('TESLA Prices Prediction')
plt.xlabel('Date')
plt.ylabel('Prices')
plt.xticks(np.arange(881,1259,50), df.Date[881:1259:50])
plt.legend()
plt.show()
• ARIMA model results in appreciable results. This model offers a
good prediction accuracy and to be relatively fast compared to
other alternatives, in terms of training/fitting time and
complexity.
• ARIMA /GARCH Vs traditional BUY/HOLD
• As you can see, over a 65 year period, the ARIMA+GARCH
strategy has significantly outperformed "Buy & Hold". However,
you can also see that the majority of the gain occured between
1970 and 1980. Notice that the volatility of the curve is quite
minimal until the early 80s, at which point the volatility increases
significantly and the average returns are less impressive.
• First of all, let's consider the fact that the ARMA model was only
published in 1951. It wasn't really widely utilised until the 1970's
when Box & Jenkins discussed it in their book.
• Secondly, the ARCH model wasn't discovered (publicly!) until
the early 80s, by Engle, and GARCH itself was published by
Bollerslev in 1986.
• Thirdly, this "backtest" has actually been carried out on a stock
market index and not a physically tradeable instrument.
• Equity curve of ARIMA+GARCH strategy vs "Buy & Hold"
for the S&P500 from 2005 until today
• As you can see the equity curve remains below a Buy & Hold
strategy for almost 3 years, but during the stock market crash of
2008/2009 it does exceedingly well. This makes sense because
there is likely to be a significant serial correlation in this period
and it will be well-captured by the ARIMA and GARCH models.
Once the market recovered post-2009 and enters what looks to
be more a stochastic trend, the model performance begins to
suffer once again.
• Note that this strategy can be easily applied to different stock
market indices, equities or other asset classes. I strongly
encourage you to try researching other instruments, as you may
obtain substantial improvements on the results presented here.
Options, Swaps, Hedgeing
and Math
03-04-2021
Last Week….
• BIG SETBACK! PPF, NSC, • Why?
Sukanya Samriddhi, Senior
Citizens Savings Scheme
interest rates SLASHED;
check new rates
here…headlines from
popular dailies…
options
Options are financial instruments that are derivatives based on
the value of underlying securities such as stocks. An options
contract offers the buyer the opportunity to buy or sell—
depending on the type of contract they hold—the underlying
asset. Unlike futures the holder is not required to buy or sell the
asset if they choose not to.
• Call options allow the holder to buy the asset at a stated price
within a specific timeframe
• Put options allow the holder to sell the asset at a stated price
within a specific timeframe
• Each option contract will have a specific expiration date by
which the holder must exercise their option. The stated price on
an option is known as the strike price. Options are typically
bought and sold through online or retail brokers
Options are a versatile financial product. These contracts involve
a buyer and a seller, where the buyer pays an
options premium for the rights granted by the contract. Each call
option has a bullish buyer and a bearish seller, while put options
have a bearish buyer and a bullish seller.
• Options Contracts usually represent 100 shares of the
underlying security, and the buyer will pay a premium fee for
each contract. For example, if an option has a premium of 35
cents per contract, buying one option would cost $35 ($0.35 x
100 = $35). The premium is partially based on the strike price—
the price for buying or selling the security until the expiration
date. Another factor in the premium price is the expiration date.
Just like with that carton of milk in the refrigerator, the expiration
date indicates the day the option contract must be used. The
underlying asset will determine the use-by date. For stocks, it is
usually the last Thursday of the contract's month.
Traders and investors will buy and sell options for several
reasons. Options speculation allows a trader to hold a leveraged
position in an asset at a lower cost than buying shares of the
asset. Investors will use options to hedge or reduce the risk
exposure of their portfolio. In some cases, the option holder
can generate income when they buy call options or become an
options writer. Options are also one of the most direct ways
to invest in oil. For options traders, an option's daily
trading volume and open interest are the two key numbers to
watch in order to make the most well-informed investment
decisions
American options can be exercised any time before the
expiration date of the option, while European options can only be
exercised on the expiration date or the exercise date. Exercising
means utilizing the right to buy or sell the underlying
security.India is the European model
Options Risk Metrics: The Greeks
• The “Greeks" is a term used in the options market to describe
the different dimensions of risk involved in taking an options
position, either in a particular option or a portfolio of options.
These variables are called Greeks because they are typically
associated with Greek symbols. Each risk variable is a result of
an imperfect assumption or relationship of the option with
another underlying variable. Traders use different Greek values,
such as delta, theta, and others, to assess options risk and
manage option portfolios.
Delta
Delta(Δ) represents the rate of change between the option's price
and a $1 change in the undelying asset’s price. In other words,
the price sensitivity of the option relative to the underlying. Delta
of a call option has a range between zero and one, while the
delta of a put option has a range between zero and negative one.
For example, assume an investor is long a call option with a delta
of 0.50. Therefore, if the underlying stock increases by $1, the
option's price would theoretically increase by 50 cents.
• For options traders, delta also represents the hedge ratio for
creating a delta neutral position. For example if you purchase a
standard American call option with a 0.40 delta, you will need to
sell 40 shares of stock to be fully hedged. Net delta for a
portfolio of options can also be used to obtain the portfolio's
hedge ration.
• A less common use of an option’s delta is it's current probability
that it will expire in-the money. For instance, a 0.40 delta call
option today has an implied 40% probability of finishing in-the-
money.
Theta
• Theta (Θ) represents the rate of change between the option price
and time, or time sensitivity - sometimes known as an option's time
decay. Theta indicates the amount an option's price would decrease
as the time to expiration decreases, all else equal. For example,
assume an investor is long an option with a theta of -0.50. The
option's price would decrease by 50 cents every day that passes, all
else being equal. If three trading days pass, the option's value would
theoretically decrease by $1.50.
• Theta increases when options are at-the-money, and decreases
when options are in- and out-of-the money. Options closer to
expiration also have accelerating time decay. Long calls and long
puts will usually have negative Theta; short calls and short puts will
have positive Theta. By comparison, an instrument whose value is
not eroded by time, such as a stock, would have zero Theta.
Gamma
• Gamma (Γ) represents the rate of change between an
option’s delta and the underlying asset's price. This is called
second-order (second-derivative) price sensitivity. Gamma
indicates the amount the delta would change given a $1 move
in the underlying security. For example, assume an investor is
long one call option on hypothetical stock XYZ. The call option
has a delta of 0.50 and a gamma of 0.10. Therefore, if stock
XYZ increases or decreases by $1, the call option's delta would
increase or decrease by 0.10.
• Gamma is used to determine how stable an option's delta is: higher
gamma values indicate that delta could change dramatically in
response to even small movements in the underlying's price.
Gamma is higher for options that are at-the-money and lower for
options that are in- and out-of-the-money, and accelerates in
magnitude as expiration approaches. Gamma values are generally
smaller the further away from the date of expiration; options with
longer expirations are less sensitive to delta changes. As expiration
approaches, gamma values are typically larger, as price changes
have more impact on gamma.
• Options traders may opt to not only hedge delta but also gamma in
order to be delta-gamma-neutral, meaning that as the underlying
price moves, the delta will remain close to zero.
Vega

• Vega (V) represents the rate of change between an option's

value and the underlying asset’s implied volatility . This is the
option's sensitivity to volatility. Vega indicates the amount an
option's price changes given a 1% change in implied volatility.
For example, an option with a Vega of 0.10 indicates the
option's value is expected to change by 10 cents if the implied
volatility changes by 1%.
• Because increased volatility implies that the underlying
instrument is more likely to experience extreme values, a rise in
volatility will correspondingly increase the value of an option.
Conversely, a decrease in volatility will negatively affect the
value of the option. Vega is at its maximum for at-the-money
options that have longer times until expiration.
Rho
• Rho (p) represents the rate of change between an option's
value and a 1% change in the interest rate. This measures
sensitivity to the interest rate. For example, assume a call
option has a rho of 0.05 and a price of $1.25. If interest rates
rise by 1%, the value of the call option would increase to $1.30,
all else being equal. The opposite is true for put options. Rho is
greatest for at-the-money options with long times until
expiration.
• Minor Greeks
• Some other Greeks, with aren't discussed as often,
are lambda,epsilon, vomma, vera, speed, zomma, color, ultima.
• These Greeks are second- or third-derivatives of the pricing
model and affect things such as the change in delta with a
change in volatility and so on. They are increasingly used in
options trading strategies as computer software can quickly
compute and account for these complex and sometimes
esoteric risk factors.
 Risk and Profits From Buying Call
Options
• As mentioned earlier, the call options let the holder buy an underlying security at the
stated strike price by the expiration date called the expiry. The holder has no
obligation to buy the asset if they do not want to purchase the asset. The risk to the
call option buyer is limited to the premium paid. Fluctuations of the underlying stock
have no impact.
• Call options buyers are bullish on a stock and believe the share price will rise above
the strike price before the option's expiry. If the investor's bullish outlook is realized
and the stock price increases above the strike price, the investor can exercise the
option, buy the stock at the strike price, and immediately sell the stock at the current
market price for a profit.
• Their profit on this trade is the market share price less the strike share price plus the
expense of the option—the premium and any brokerage commission to place the
orders. The result would be multiplied by the number of option contracts purchased,
then multiplied by 100—assuming each contract represents 100 shares.
• However, if the underlying stock price does not move above the strike price by the
expiration date, the option expires worthlessly. The holder is not required to buy the
shares but will lose the premium paid for the call.
Risk and Profits From Selling Call
Options
• Selling call options is known as writing a contract. The writer receives the
premium fee. In other words, an option buyer will pay the premium to the writer—
or seller—of an option. The maximum profit is the premium received when selling
the option. An investor who sells a call option is bearish and believes the
underlying stock's price will fall or remain relatively close to the option's strike
price during the life of the option.
• If the prevailing market share price is at or below the strike price by expiry, the
option expires worthlessly for the call buyer. The option seller pockets the
premium as their profit. The option is not exercised because the option buyer
would not buy the stock at the strike price higher than or equal to the prevailing
market price.
• However, if the market share price is more than the strike price at expiry, the
seller of the option must sell the shares to an option buyer at that lower strike
price. In other words, the seller must either sell shares from their portfolio
holdings or buy the stock at the prevailing market price to sell to the call option
buyer. The contract writer incurs a loss. How large of a loss depends on the cost
basis of the shares they must use to cover the option order, plus any brokerage
order expenses, but less any premium they received.
• As you can see, the risk to the call writers is far greater than the risk exposure of
call buyers. The call buyer only loses the premium. The writer faces infinite risk
because the stock price could continue to rise increasing losses significantly.
Risk and Profits From Buying Put Options
• Put options are investments where the buyer believes the underlying stock's
market price will fall below the strike price on or before the expiration date of the
option. Once again, the holder can sell shares without the obligation to sell at the
stated strike per share price by the stated date
• Since buyers of put options want the stock price to decrease, the put option is
profitable when the underlying stock's price is below the strike price. If the
prevailing market price is less than the strike price at expiry, the investor can
exercise the put. They will sell shares at the option's higher strike price. Should
they wish to replace their holding of these shares they may buy them on the open
market.
• Their profit on this trade is the strike price less the current market price, plus
expenses—the premium and any brokerage commission to place the orders. The
result would be multiplied by the number of option contracts purchased, then
multiplied by 100—assuming each contract represents 100 shares.
• The value of holding a put option will increase as the underlying stock price
decreases. Conversely, the value of the put option declines as the stock price
increases. The risk of buying put options is limited to the loss of the premium if
the option expires worthlessly.
Risk and Profits From Selling Put Options
• Selling put options is also known as writing a contract. A put option writer believes the underlying
stock's price will stay the same or increase over the life of the option—making them bullish on the
shares. Here, the option buyer has the right to make the seller, buy shares of the underlying asset
at the strike price on expiry.
• If the underlying stock's price closes above the strike price by the expiration date, the put option
expires worthlessly. The writer's maximum profit is the premium. The option isn't exercised
because the option buyer would not sell the stock at the lower strike share price when the market
price is more.
• However, if the stock's market value falls below the option strike price, the put option writer is
obligated to buy shares of the underlying stock at the strike price. In other words, the put option
will be exercised by the option buyer. The buyer will sell their shares at the strike price since it is
higher than the stock's market value.
• The risk for the put option writer happens when the market's price falls below the strike price.
Now, at expiration, the seller is forced to purchase shares at the strike price. Depending on how
much the shares have depreciated, the put writer's loss can be significant.
• The put writer—the seller—can either hold on to the shares and hope the stock price rises back
above the purchase price or sell the shares and take the loss. However, any loss is offset
somewhat by the premium received.
• Sometimes an investor will write put options at a strike price that is where they see the shares
being a good value and would be willing to buy at that price. When the price falls, and the option
buyer exercises their option, they get the stock at the price they want, with the added benefit of
receiving the option premium.
• Suppose that Microsoft shares are trading at $108 per share
and you believe that they are going to increase in value. You
decide to buy a call option to benefit from an increase in the
stock's price.
• You purchase one call option with a strike price of $115 for one
month in the future for 37 cents per contact. Your total cash
outlay is $37 for the position, plus fees and commissions (0.37 x
100 = $37).
• If the stock rises to $116, your option will be worth $1, since you
could exercise the option to acquire the stock for $115 per share and
immediately resell it for $116 per share. The profit on the option
position would be 170.3% since you paid 37 cents and earned $1—
that's much higher than the 7.4% increase in the underlying stock
price from $108 to $116 at the time of expiry.
• In other words, the profit in dollar terms would be a net of 63 cents or
$63 since one option contract represents 100 shares [($1 - 0.37) x
100 = $63].
• If the stock fell to $100, your option would expire worthlessly, and
you would be out $37 premium. The upside is that you didn't buy 100
shares at $108, which would have resulted in an $8 per share, or
$800, total loss. As you can see, options can help limit your
downside risk.
Options Spreads

• Options spreads are strategies that use various combinations of

buying and selling different options for a desired risk-return
profile. Spreads are constructed using vanilla options, and can
take advantage of various scenarios such as high- or low-
volatility environments, up- or down-moves, or anything in-
between.
• Spread strategies, can be characterized by their payoff or
visualizations of their profit-loss profile, such as bull call
spreads /iron condors
•i
Swaps
• Derivatives contracts can be divided into two general families:
1. Contingent claims (e.g., options)
2. Forward claims, which include exchange-traded futures,
forward contracts, and swaps
A swap is an agreement between two parties to exchange
sequences of cash flows for a set period of time. Usually, at the
time the contract is initiated, at least one of these series of cash
flows is determined by a random or uncertain variable, such as
an interest rate, foreign exchange rate, equity price, or
commodity price
• Conceptually, one may view a swap as either a portfolio of
forward contracts or as a long position in one bond coupled with
a short position in another bond. The two most common and
most basic types of swaps: interest rate and currency
• In finance, a swap is a derivative contract in which one party
exchanges or swaps the values or cash flows of one asset for
another.
• Of the two cash flows, one value is fixed and one is variable
and based on an index price, interest rate, or currency
exchange rate.
• Swaps are customized contracts traded in the over-the-counter
(OTC) market privately, versus options and futures traded on a
public exchange.
• The plain vanilla interest rate and currency swaps are the two
most common and basic types of swaps.
• Unlike most standardized options and futures contracts, swaps
are not exchange-traded instruments. Instead, swaps are
customized contracts that are traded in the over-the-counter
(OTC) market between private parties
• Firms and financial institutions dominate the swaps market, with
few (if any) individuals ever participating. Because swaps occur
on the OTC market, there is always the risk of
a counterparty defaulting on the swap.
• The first interest rate swap occurred between IBM and the
World Bank in 1981.1 However, despite their relative youth,
swaps have exploded in popularity. In 1987, the International
Swaps and Derivatives Association reported that the swaps
market had a total notional value of $865.6 billion.2 By mid-
2006, this figure exceeded $250 trillion, according to the Bank
for International Settlements.3 That's more than 15 times the
size of the U.S. public equities market.
Plain Vanilla Interest Rate Swap

The most common and simplest swap is a plain vanilla interest

rate swap. In this swap, Party A agrees to pay Party B a
predetermined, fixed rate of interest on a notional principal on
specific dates for a specified period of time. Concurrently, Party B
agrees to make payments based on a floating interest rate to
Party A on that same notional principal on the same specified
dates for the same specified time period. In a plain vanilla swap,
the two cash flows are paid in the same currency. The specified
payment dates are called settlement dates, and the times
between are called settlement periods. Because swaps are
customized contracts, interest payments may be made annually,
quarterly, monthly, or at any other interval determined by the
parties.
• For example, imagine ABC Co. has just issued $1 million in five-
year bonds with a variable annual interest rate defined as
the LIBOR plus 1.3% (or 130 basis points). Also, assume that
LIBOR is at 2.5% and ABC management is anxious about an
interest rate rise.
• The management team finds another company, XYZ Inc., that is
willing to pay ABC an annual rate of LIBOR​ plus 1.3% on a
notional principal of $1 million for five years. In other words,
XYZ will fund ABC's interest payments on its latest bond
issue. In exchange, ABC pays XYZ a fixed annual rate of 5% on
a notional value of $1 million for five years. ABC benefits from
the swap if rates rise significantly over the next five years. XYZ
benefits if rates fall, stay flat, or rise only gradually.
• LIBOR, or London Interbank Offered rate is the interest rate
offered by London banks on deposits made by other banks in
the Eurodollar markets. The market for interest rate swaps
frequently (but not always) used LIBOR as the base for the
floating rate until 2020. The transition from LIBOR to other
benchmarks, such as the secured overnight financing
rate (SOFR), began in 2020.
Scenario 1
If LIBOR rises by 0.75% per year, Company ABC's total interest
payments to its bondholders over the five-year period amount to
$225,000. Let's break down the calculation:
 Libor + 1.30% Variable Interest 5% Interest Paid ABC's Gain XYZ's Loss
Paid by XYZ to by ABC to XYZ
ABC
Year 1 3.80% $38,000 $50,000 -$12,000 $12,000
Year 2 4.55% $45,500 $50,000 -$4,500 $4,500
Year 3 5.30% $53,000 $50,000 $3,000 -$3,000
Year 4 6.05% $60,500 $50,000 $10,500 -$10,500
Year 5 6.80% $68,000 $50,000 $18,000 -$18,000
Total $15,000 ($15,000)
• In this scenario, ABC did well because its interest rate was fixed
at 5% through the swap. ABC paid $15,000 less than it would
have with the variable rate. XYZ's forecast was incorrect, and
the company lost $15,000 through the swap because rates rose
faster than it had expected.
• If LIBOR rises 0,25% per year
Scenario 2
In the second scenario, LIBOR rises by 0.25% per year:

Variable
5% Interest
Interest Paid
Libor + 1.30% Paid by ABC ABC's Gain XYZ's Loss
by XYZ to
to XYZ
ABC
Year 1 3.80% $38,000 $50,000 ($12,000) $12,000
Year 2 4.05% $40,500 $50,000 ($9,500) $9,500
Year 3 4.30% $43,000 $50,000 ($7,000) $7,000
Year 4 4.55% $45,500 $50,000 ($4,500) $4,500
Year 5 4.80% $48,000 $50,000 ($2,000) $2,000
Total ($35,000) $35,000
• In this case, ABC would have been better off by not engaging in the
swap because interest rates rose slowly. XYZ profited $35,000 by
engaging in the swap because its forecast was correct.
• This example does not account for the other benefits ABC might
have received by engaging in the swap. For example, perhaps the
company needed another loan, but lenders were unwilling to do that
unless the interest obligations on its other bonds were fixed.
• In most cases, the two parties would act through a bank or other
intermediary, which would take a cut of the swap. Whether it is
advantageous for two entities to enter into an interest rate swap
depends on their comparative advantage in fixed or floating-
rate lending markets.
• The instruments exchanged in a swap do not have to be
interest payments. Countless varieties of exotic swap
agreements exist, but relatively common arrangements include
commodity swaps, currency swaps, debt swaps, and total return
swaps.
Commodity Swaps

Commodity swaps involve the exchange of a floating commodity

price, such as the Brent Crude oil spot price, for a set price over
an agreed-upon period. As this example suggests, commodity
swaps most commonly involve crude oil.
• In a currency swap, the parties exchange interest and principal
payments on debt denominated in different currencies. Unlike
an interest rate swap, the principal is not a notional amount, but
it is exchanged along with interest obligations. Currency swaps
can take place between countries. For example, China has
used swaps with Argentina, helping the latter stabilize its foreign
reserves.The US Federal Reserve engaged in an aggressive
swap strategy with European central banks during the 2010
European financial crisis to stabilize the euro, which was falling
in value due to the Greek debt crisis
Debt-Equity Swap
• A debt-equity swap involves the exchange of debt for equity—in
the case of a publicly-traded company, this would mean bonds
for stocks. It is a way for companies to refinance their debt or
reallocate their capital structure.
Total Return Swap
• In a total return swap, the total return from an asset is
exchanged for a fixed interest rate. This gives the party paying
the fixed-rate exposure to the underlying asset—a stock or
an index. For example, an investor could pay a fixed rate to one
party in return for the capital appreciation
plus dividend payments of a pool of stocks.
Credit Default Swap
• A (CDS) consists of an agreement by one party to pay the lost
principal and interest of a loan to the CDS buyer if a
borrower defaults on a loan. Excessive leverage and poor risk
management in the CDS market were contributing causes of
the 2008 financial crisis
Swaps Summary

• A financial swap is a derivative contract where one party

exchanges or "swaps" the cash flows or value of one asset for
another. For example, a company paying a variable rate of
interest may swap its interest payments with another company
that will then pay the first company a fixed rate. Swaps can also
be used to exchange other kinds of value or risk like the
potential for a credit default in a bond.
We shall further see
• Hedging(Hedge Funds)- Futures
• Greeks & Math –Option
• Swaps we will not be referring to as they are not tradeable
Hedging
• Index Funds
• Beta of a fund
• Why Hedging
Index
• A market index is a hypothetical portfolio of investment holdings
that represents a segment of the financial market. The
calculation of the index value comes from the prices of the
underlying holdings. Some indexes have values based on
market-cap weighting, revenue-weighting, float-weighting, and
fundamental-weighting. Weighting is a method of adjusting the
individual impact of items in an index.
• NASDAQ100, S&P500, NIFTY etc
A market index measures the value of a portfolio of holdings with
specific market characteristics. Each index has its own
methodology which is calculated and maintained by the index
provider. Index methodologies will typically be weighted by either
price or market cap. A wide variety of investors use market
indexes for following the financial markets and managing their
investment portfolios. Indexes are deeply entrenched in the
investment management business with funds using them as
benchmarks for performance comparisons and managers using
them as the basis for creating investable index funds.
Investors may choose to build a portfolio with diversified
exposure to several indexes or individual holdings from a variety
of indexes. They may also use benchmark values and
performance to follow investments by segment. Some investors
will allocate their investment portfolios based on the returns or
expected returns of certain segments. Further, a specific index
may act as a benchmark for a portfolio or a mutual fund
• Institutional fund managers use benchmarks as a proxy for a fund’s
individual performance. Each fund has a benchmark discussed in
its prospectus and provided in its performance reporting, thus offering
transparency to investors. Fund benchmarks can also be used to evaluate
the compensation and performance of fund managers.
• Institutional fund managers also use indexes as a basis for creating index
funds. Individual investors cannot invest in an index without buying each of
the individual holdings, which is generally too expensive from a trading
perspective. Therefore, index funds are offered as a low-cost way for
investors to invest in a comprehensive index portfolio, gaining exposure to
a specificmarket segment of their choosing. Index funds use an index
replication strategy that buys and holds all of the constituents in an index.
Some management and trading costs are still included in the
fund’s expense ratio, but the costs are much lower than fees for an actively
managed fund.
• The NIFTY 50 is a diversified 50 stock index accounting for 13 sectors of
the economy. It is used for a variety of purposes such as benchmarking
fund portfolios, index based derivatives and index funds.
• NIFTY 50 is owned and managed by NSE Indices Limited (formerly known
as India Index Services & Products Limited) (NSE Indices). NSE Indices is
India's specialised company focused upon the index as a core product.
• The NIFTY 50 Index represents about 66.8% of the free float market
capitalization of the stocks listed on NSE as on March 29, 2019.
• The total traded value of NIFTY 50 index constituents for the last six
months ending March 2019 is approximately 53.4% of the traded value of
all stocks on the NSE.
• Impact cost of the NIFTY 50 for a portfolio size of Rs.50 lakhs is 0.02% for
the month March 2019..
• NIFTY 50 is ideal for derivatives trading.
Some terms….
• Beta is the return generated from a portfolio that can be attributed to
overall market returns. Exposure to beta is equivalent to exposure to
systematic risk.
• Alpha is the portion of a portfolio's return that cannot be attributed to
market returns and is thus independent of them. Exposure to alpha is
equivalent to exposure to idiosyncratic risk.
• Systematic Risk is the risk that comes from investing in any security
within the market. The level of systematic risk that individual security
possesses depends on how correlated it is with the overall market. This is
quantitatively represented by beta exposure.
• Idiosyncratic Risk is the risk that comes from investing in single security
(or investment class). The level of idiosyncratic risk individual security
possesses is highly dependent on its own unique characteristics. This is
quantitatively represented by alpha exposure. (Note: A single alpha
position has its own idiosyncratic risk. When a portfolio contains more than
one alpha position, the portfolio will then reflect each alpha position's
idiosyncratic risk collectively.)
• Separating a single portfolio into two portfolios --
an alpha portfolio and a beta portfolio -- affords an investor
greater control over the entire combination of exposure risks. By
individually selecting your exposure to alpha and beta, you can
enhance returns by consistently maintaining desired risk levels
within your aggregate portfolio
• This measurement of portfolio returns is called the alpha-beta
framework. An equation is derived with
linear regression analysis by using the portfolio's return
compared to the return of the market over the same period of
time. The equation calculated from the regression analysis will
be a simple line equation that "best fits" the data. The slope of
the line produced from this equation is the portfolio's beta, and
the y-intercept (the part that cannot be explained by market
returns) is the alpha that was generated.
The Beta Exposure
• A portfolio that is constructed of multiple equities will inherently
have some beta exposure. Beta exposure in individual security
is not a fixed value over a given period of time. This translates
to systematic risk that cannot be held at a steady value. By
separating the beta component, an investor can keep a
controlled set amount of beta exposure in accordance with their
own risk tolerance. This helps enhance portfolio returns by
producing more consistent portfolio returns.
Alpha and beta expose portfolios to idiosyncratic risk and
systematic risk, respectively; however, this is not necessarily a
negative thing. The degree of risk to which an investor is exposed
is correlated to the degree of potential return that can be
expected.
• Before you can choose a level of beta exposure, you must first
choose an index that you feel represents the overall market.
• After selecting an index, you must choose the desired level of
beta exposure for your portfolio. If you invest 50% of your
capital in an Index fund and keep the rest in cash, your portfolio
has a beta of 0.5. If you invest 70% of your capital in an Index
fund and keep the rest in cash, your portfolio beta is 0.7. This is
because the represents the overall market, which means that it
will have a beta of 1.
• Why?
• If V =Current Value of portfolio
• If VF=Current Value of Future Contracts

N= V/VF where N is no of contracts to go long or short. We know the

value from NIFTY futures
Beta Exposure
• Buy indexed fund
• Buy futures
• Buy a combination
Alpha
• For an investment to be considered pure alpha, its returns must
be completely independent of the returns attributed to beta.
Some strategies that exemplify the definition of pure alpha
are stastical arbitrage(quant or stat trading), equity neutral
hedged strategies and selling liquidity premiums in the fixed-
income market.
• Some portfolio managers use their alpha portfolios to buy
individual equities. This method is not pure alpha, but rather the
manager's skill in equity selection. This creates a positive alpha
return, but it is what is referred to as “tainted alpha”It is tainted
because of the consequential beta exposure that goes along
with the purchase of the individual equity, which keeps this
return from being a pure alpha.
• Individual investors trying to replicate this strategy will find the
latter scenario of producing tainted alpha to be the preferred
method of execution. This is due to the inability to invest in the
professionally run, privately owned funds (casually called hedge
funds) that specialize in pure alpha strategies.
• Alfred Winslow-1949-Father of the Hedge fund
• Stock picking and hedging
Capital Asset Pricing Model (CAPM)

The formula for calculating expected rate of return given the risk
ERi​=Rf​+βi​(ERm​−Rf​)
• where:ERi​=expected return of investment
• Rf​=riskfree rate
• βi​=beta of the investment
• (ERm​−Rf​)=market risk premium​
• The beta of a potential investment is a measure of how much
risk the investment will add to a portfolio that looks like the
market. If a stock is riskier than the market, it will have a beta
greater than one. If a stock has a beta of less than one, the
formula assumes it will reduce the risk of a portfolio.
Adjusting a Stock Portfolio’s Beta using
Stock Index Futures
• Beta, as defined in the capital asset pricing model, is a measure
of a portfolio’s systematic risk. When a trader uses index futures
to hedge a position in an equity portfolio, they are effectively
trying to reduce the portfolio’s systematic risk. As such, hedging
is actually an attempt to reduce a portfolio’s beta.
Let’s define the following variables:
Β=portfolio beta
β∗=target beta after hedging
P=portfolio value
A=Unit value of the underlying Asset
Number of contracts required=(β∗−β)(P/A)
If the above result is positive, the trader would have to buy the
relevant number of futures. If negative, they would have to sell
futures.
Efficient
Frontier
Two portfolios that have
been constructed to fit
along the efficient frontier.
Portfolio A is expected to
return 8% per year and
has a 10% standard
deviation or risk level.
Portfolio B is expected to
return 10% per year but
has a 16% standard
deviation. The risk of
portfolio B rose faster than
its expected returns.
Security Market Line

• This trade-off between risk

and return applies to the
CAPM and the efficient frontier
graph can be rearranged to
illustrate the trade-off for
individual assets. In the
following chart, you can see
that the CML is now called
the Security Market Line
Instead of expected risk on the
x-axis, the stock’s beta is
used. As you can see in the
illustration, as beta increases
from one to two, the expected
return is also rising
Math on Options
 Delta (Δ) is a measure of the sensitivity of an option’s price changes relative to
the changes in the underlying asset’s price. In other words, if the price of the
underlying asset increases by $1, the price of the option will change by Δ
amount. Mathematically, the delta is found by:

Where:
•∂ – the first derivative
•V – the option’s price (theoretical value)
•S – the underlying asset’s price

Keep in mind that we take the first derivative of the option and underlying asset
prices because the derivative is a measure of the rate change of the variable
Delta at a given period.
The delta is usually calculated as a decimal number from -1 to 1. Call
options can have a delta from 0 to 1, while puts have a delta from -1 to 0. The
closer the option’s delta to 1 or -1, the deeper in-the-money is the option.
The delta of an option’s portfolio is the weighted average of the deltas of all
options in the portfolio.
Delta is also known as a hedge ratio. If a trader knows the delta of the option,
he can hedge his position by buying or shorting the number of underlying
assets multiplied by delta.
• Delta option greeks are the most popular of the Greeks
because it’s the easiest to understand. It measures the rate of
change in price. In other words, Delta tells us how much an
option would increase when the stock moves up a point.
• It’ll be positive for call options and negative for put options.
Depending on your strategy, you can use Delta to determine the
probability of the contract being in the money at expiration.
• Many times, you want to be at the money at expiration because
the more in the money you are, the higher the your profit
potential is. Many times when you go to place and options
trade, you want a higher delta.
• he higher the delta the higher the risk and reward. While there
is high risk, you can offset that by practicing proper risk
management and trading patterns. You want traction so where
you buy your option determines how well your contract moves.
• Remember that the higher the delta the more expensive the
option will be. The reason is because you have a higher
probability of expiring in the money. Hence a higher profit
potential. In the money gives you more of a cushion too.
 Gamma (Γ) is a measure of the
delta’s change relative to the
changes in the price of the
underlying asset. If the price of the
Gamma underlying asset increases by $1,
the option’s delta will change by the
gamma amount. The main
application of gamma is the
assessment of the option’s delta.

Long options have a positive

gamma. An option has a maximum
gamma when it is at-the-money
(option strike price equals the price
of the underlying asset). However,
gamma decreases when an option is
deep-in-the-money or out-the-money.
• Gamma option greeks are the most ambiguous of the option Greeks.
In essence it measures sensitivity to Delta. Delta increases and
decreases with price movement so Gamma stays constant to
measure that change. The higher the Gamma the more risky the
trade. The lower the Gamma the less risky. The risk you want to
incur is up to you. Many times the riskier the trade, when payed right,
is a lot more rewarding.
• A higher Gamma means more volatility. That’s something you have
to be able to stomach. Especially for longer term options trades.
Which means direction matters.
• Hence playing candlesticks, patterns and trends will help
immensely.
Vega
Vega (ν) is an option Greek that measures the sensitivity
of an option price relative to the volatility of the
underlying asset. If the volatility of the underlying asses
increases by 1%, the option price will change by the vega
amount.

Where:
•∂ – the first derivative
•V – the option’s price (theoretical value)
•σ – the volatility of the underlying asset

The vega is expressed as a money amount rather than as

a decimal number. An increase in vega generally
corresponds to an increase in the option value (both calls
and puts).
• Vega makes up another of the option Greeks. This options Greek
maps out the sensitivity of volatility. It doesn’t matter if you’re trading
stocks or options, volatility will always be important.
• Vega measures the rate of change in implied volatility. Implied
volatility is the expected volatility of an option. In other words, Vega
shows you how much an option’s price will change for every 1%
implied volatility moves.
• You can use Vega to see how different strategies would work when
they look the same based off the Delta. Hence option Greeks go
hand in hand. You can’t really have one without the other.
 Theta (θ) is a measure of the sensitivity of
the option price relative to the option’s time
to maturity. If the option’s time to maturity
decreases by one day, the option’s price will
change by the theta amount. The Theta

Theta option Greek is also referred to as time

decay.

Where:
•∂ – the first derivative
•V – the option’s price (theoretical value)
•τ – the option’s time to maturity
In most cases, theta is negative for options.
However, it may be positive for some
European options. Theta shows the most
negative amount when the option is at-the-
money.
• Theta deals with time decay. Options are wasting assets
because they expire. Time is constantly moves forward. As a
result, your options decay. Think of it like an hourglass.
• The more time passes, the seller of the option profits. So the
seller would be the bottom of the hourglass whereas the buyer
would be at the top. As a result, profit drips to the seller on a
daily basis.
• 80% of options expire worthless. That means many times, the
seller is the winner.
Black Scholes model
• The Black-Scholes model, also known as the Black-Scholes-
Merton (BSM) model, is a mathematical model for pricing an
options contract. In particular, the model estimates the variation
over time of financial instruments.
• The Black-Scholes model is one of the most important concepts
in modern financial theory. It was developed in 1973 by Fischer
Black, Robert Merton, and Myron Scholes and is still widely
used today. It is regarded as one of the best ways of
determining the fair price of options. The Black-Scholes model
requires five input variables: the strike price of an option, the
current stock price, the time to expiration, the risk-free rate, and
the volatility.
• Also called Black-Scholes-Merton (BSM), it was the first widely
used model for option pricing. It's used to calculate the
theoretical value of options using current stock prices, expected
dividends, the option's strike price, expected interest rates, time
to expiration and expected volatility.
• The initial equation was introduced in Black and Scholes' 1973
paper, "The Pricing of Options and Corporate Liabilities,"
published in the Journal of Political Economy.
• Black passed away two years before Scholes and Merton were
awarded the 1997 Nobel Prize in economics for their work in
finding a new method to determine the value of derivatives (the
Nobel Prize is not given posthumously; however, the Nobel
committee acknowledged Black's role in the Black-Scholes
model).
Black-Scholes posits that instruments, such as stock shares or
futures contracts, will have a lognormal distribution of prices
following a random walk with constant drift and volatility. Using
this assumption and factoring in other important variables, the
equation derives the price of a European-style call option.
The inputs for the Black-Scholes equation are volatality the price
of the underlying asset, the strike price of the option, the time
until expiration of the option, and the risk-free interest rate. With
these variables, it is theoretically possible for options sellers to
set rational prices for the options that they are selling.
Furthermore, the model predicts that the price of heavily traded
assets follows a geometric Brownian motion with constant drift
and volatility. When applied to a stock option, the model
incorporates the constant price variation of the stock, the time
value of money, the option's strike price, and the time to the
option's expiry.