Maladex

Research-Driven Cardano DEX

White Paper v1

Jarek Hirniak
21 October 2021

Abstract

Maladex’s goal is to revolutionize the Decentralized Finance (DeFi) ecosystem

by application of research rigour and quantitative modelling expertise from Tradi-
tional Finance (TradFi). In doing so, we are going to make the entire ecosystem
more efficient and provide novel revenue generating streams. We propose a rev-
olutionary execution model based on the concept of programmable swaps. The
proposed approach completely eradicates impermanent loss, significantly increases
capital efficiency, and is poised to render market making and price discovery much
more efficient.
Building on the foundation of programmable swaps and on-chain autonomous
code, we’re developing a platform which enables a multitude of alpha generating
products. These products could include passive investments such as crypto indexes,
through synthetics, and financial derivatives, to programmable trading strategies
akin to those employed by top performing market makers and hedge funds in tra-
ditional finance.
The platform not only connects investors with a wide range of investment prod-
ucts, but does so while enabling risk control, dynamic hedging, and adaptability
to market conditions. Maladex intends to provide high-frequency on-chain data
streams, data warehouse and tooling for backtesting. Finally, we intend to engage
in research and education to move DeFi forward.
The Maladex platform is going to decentralize and democratize access to be-
spoke trading tools and investments for the entire world.
The platform will use a decentralized liquidity provisioning protocol which forms
the foundation on which highly efficient blockchain based crypto markets can rest.
It enables protocol participants to create a variety of financial products from index
funds, options, synthetics, hedge funds all the way to insurance products.
 In this document, we introduce all the concepts outlined above and the intrica-
cies of financial markets. We invite you on the journey with us via this white paper
and the tools we are going to build.

2
1 Disclaimer
Cardano is a new, third generation blockchain, with a novel Extended Unspent Transac-
tion Output (EUTxO) model and the Plutus smart contract programming language under
active development. Maladex is a platform proposing a series of innovative and revolution-
ary concepts. Therefore, matters described in this white paper are subject to potential
change in the future, carry unknown risks that might impact parts of the project, and
create opportunities for new discoveries that could require rethinking some of the initial
assumptions. We reserve the rights to adjust the plan. Any new endeavours are filled
with discoveries and novel ideas, and we plan to capitalize on them and adjust to them.
Below is the vision of the Maladex platform in the long term and does not represent a
feature list planned for launch.

3
Contents
1 Disclaimer 3

2 Introduction 9
2.1 Market participants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Liquidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 Efficient Market Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.4 Automated Market Makers & Decentralized Exchanges . . . . . . . . . . 12
2.5 Constant-Function Market Makers . . . . . . . . . . . . . . . . . . . . . . 12
2.6 Concentrated Liquidity . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.7 Capital Efficiency ξ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.8 Impermanent Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.9 Risk Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.10 Cardano Blockchain & EUTxO Model . . . . . . . . . . . . . . . . . . . 17
2.11 What’s Next . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

3 Maladex Protocol Mission & Objectives 18

3.1 Providing the Building Blocks of Sound and Efficient Ways of Investing to
Everyone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.2 Increase Market Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.3 Build the Foundations for the Future Financial Markets . . . . . . . . . . 19

4 Programmable Swaps 20
4.1 Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 Commit Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.3 Execution Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.4 Maladex Execution Pipeline . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.5 Scalability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.6 Memory Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.7 EOL of Programmable Swaps . . . . . . . . . . . . . . . . . . . . . . . . 26
4.7.1 NFT Receipt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.8 Theoretical Model Performance Characteristics . . . . . . . . . . . . . . . 27
4.9 Fragments Matching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.9.1 Market Order Matched by Limit Order . . . . . . . . . . . . . . . 28
4.9.2 Market Order Matched by AAMM -LP . . . . . . . . . . . . . . . 30
4.9.3 Limit Order Matched by the Order Book . . . . . . . . . . . . . . 32
4.10 Programmable Swaps Domain-Specific Language . . . . . . . . . . . . . . 35

4
 4.11 User Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.12 Fundamental Building Blocks . . . . . . . . . . . . . . . . . . . . . . . . 36

5 Maladex Order Matching Engine 38

5.1 Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.2 Routing Rules and Fairness . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.3 Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.4 Scalability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.5 Decentralization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

6 Algorithmic Automated Market Making (AAMM) 40

6.1 Impermanent Loss & Constant-Function Market Maker . . . . . . . . . . 40
6.2 Reducing Impact of Impermanent Loss . . . . . . . . . . . . . . . . . . . 42
6.3 Programmable Source and Target Asset Ratios . . . . . . . . . . . . . . . 43
6.4 Realistic Liquidity Supply Curve . . . . . . . . . . . . . . . . . . . . . . 44
6.5 Algorithmic Liquidity Pool . . . . . . . . . . . . . . . . . . . . . . . . . . 46
6.6 Offsetting Internal Risk with External Liquidity . . . . . . . . . . . . . . 47
6.7 Minting Sound Liquidity Pools . . . . . . . . . . . . . . . . . . . . . . . . 47
6.8 Fragmented Liquidity of AAMM . . . . . . . . . . . . . . . . . . . . . . . 48

7 Yield Curve 49
7.1 Yield Farming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
7.2 ADA Staking Rewards from Smart Contracts . . . . . . . . . . . . . . . . 49

8 Emerging Properties of Programmable Swaps and AAMM 50

9 Indexes 51
9.1 Index Balancing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
9.2 Index Categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
9.3 Cryptocurrency Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

10 Synthetics 53
10.1 Advantages of Using Synthetics . . . . . . . . . . . . . . . . . . . . . . . 53
10.2 Synthetic Token Interface . . . . . . . . . . . . . . . . . . . . . . . . . . 54

11 Financial Derivatives 55
11.1 Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
11.2 Advantages of option trading on the blockchain . . . . . . . . . . . . . . 56
11.3 Option Trading Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . 57

5
 11.3.1 Bull Call Spread . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
11.3.2 Bear Put Spread . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
11.3.3 Long Straddle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
11.3.4 Long Strangle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
11.3.5 Other Option Strategies . . . . . . . . . . . . . . . . . . . . . . . 59

12 Arbitrage 60

13 Oracles 61

14 Risk Control 62
14.1 Why do people take risk? . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
14.2 Market Maker Risk Compensation . . . . . . . . . . . . . . . . . . . . . . 62
14.3 Dynamic Risk Compensation . . . . . . . . . . . . . . . . . . . . . . . . 62
14.4 Token Trust Scores & Whistleblowing . . . . . . . . . . . . . . . . . . . . 63
14.5 Quantification of Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
14.6 Moment Indicators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

15 Maladex Protocol Settlement Layer 65

15.1 Maladex as Scaling Layer for Cardano-native Projects . . . . . . . . . . . 65

16 High-Frequency Data Lake & Lab 66

17 On-chain Hedge Fund 67

17.1 On-Chain Portfolio Managers . . . . . . . . . . . . . . . . . . . . . . . . 67

18 DeFi Education Portal 68

19 Tokenomics 69
19.1 MAL Token Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
19.2 MAL Token Vesting Schedule . . . . . . . . . . . . . . . . . . . . . . . . . 70
19.3 MAL Token Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
19.4 Maladex Treasury . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Acronyms 72

Glossary 73

6
List of Figures
1 Constant product Automated Market Maker (AMM) formula (Uniswap’s
model). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2 Constant mean AMM formula (Balancer’s model). . . . . . . . . . . . . . 13
3 Hybrid Constant-Function Market Maker (CFMM) of Curve’s stableswap
vs constant-product CFMM comparison. . . . . . . . . . . . . . . . . . . 14
4 Convergence onto geometric price distribution of concentrated liquidity
model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
5 Percentage Divergence Impermanent Loss (Duplicate of Figure 17 for Eas-
ier Reading). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
6 Maladex programmable swap protocol execution stages. . . . . . . . . . . 22
7 Active and inactive frontiers. . . . . . . . . . . . . . . . . . . . . . . . . . 24
8 Maladex protocol (off-chain order matching engine) execution pipeline. . . 25
9 Fragmented (virtual) liquidity pool’s initial state S0 . . . . . . . . . . . . . 28
10 Transaction executing M0 using LS,1 . . . . . . . . . . . . . . . . . . . . . 29
11 State of the fragmented liquidity pool after executing T0 . . . . . . . . . . 29
12 Transaction executing M1 using A3 . . . . . . . . . . . . . . . . . . . . . . 31
13 State of the fragmented liquidity pool after executing T1 . . . . . . . . . . 31
14 Order book crossover. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
15 Transaction matching 3 order book transactions: LB,1 , LB,2 , and LS,1 . . . 34
16 State of the fragmented liquidity pool after executing T2 . . . . . . . . . . 34
17 Percentage Divergence of Impermanent Loss on Initial Investment of $1, 000.00
at valuation of $2.00 and 1 Token X at valuation of $1, 000.00. . . . . . . 41
18 Divergence of Impermanent Loss in USD on Initial Investment of $1, 000.00
at valuation of $2.00 and 1 Token X at valuation of $1, 000.00. . . . . . . 41
19 Liquidity Provision Asset Ratios Impact on Impermanent Loss . . . . . . 43
20 Stochastic Price Model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
21 Stochastic Process Reversal to the Mean. . . . . . . . . . . . . . . . . . . 45
22 Bull call spread with strike prices K1 (low strike price) and K2 (high strike
price). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
23 Bear put spread with strike prices K1 (low strike price) and K2 (high strike
price). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
24 Long straddle with strike price K. . . . . . . . . . . . . . . . . . . . . . . 58
25 Long strangle with strike price K. . . . . . . . . . . . . . . . . . . . . . . 59
26 MAL token allocation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
27 Plutus Application Back-end schematic. . . . . . . . . . . . . . . . . . . 79

7
List of Tables
1 Exchange Amount of Transactions per Day, Week, and Month. . . . . . . 65

8
Maladex: Research-Driven Cardano DEX 9 of 88

2 Introduction
In this section, we lay the foundations for the rest of the paper, introducing fundamental
financial engineering concepts, with a focus on their application in DeFi. We firmly
believe that DeFi is a young ecosystem that can benefit in a significant way from the
application of a wealth of existing knowledge in Quantitative Finance (QuantFi) and the
discovery of financial engineering concepts unique to the nature of DeFi and blockchain.
Let’s start at the beginning – agents who create the market.

2.1 Market participants

In the context of exchanges, market participants are classified into:

• (market) takers – are agents who wish to exchange assets. They take liquidity away
from the market by exchanging one asset for the other. They need the market to be
liquid in order to guarantee that they can both exchange the assets (immediately)
and at the same time that the asset price is not significantly affected by the mere
fact of the exchange.
• (market) makers – are agents who provide liquidity to the market. In TradFi, mar-
ket makers profit from excellent market price predictions and providing liquidity at
small profits (e.g., a cent on each share). When applied to as many market partici-
pants as they do, it guarantees them a significant over day profit. In DeFi, this role
is often deferred to the algorithm that models the liquidity curve, automatically
updates the price, and distributes market maker rewards to the liquidity providers.

As demonstrated above, both parties are fundamental for exchange operations to function.
On the one hand, without market makers, market takers would often be forced to overpay
for the assets and be limited to waiting for a party willing to take the opposite side of
the trade. On the other hand, without market takers, market makers would not be able
to generate a steady profit.

2.2 Liquidity
We have seen in 2.1 the crucial role that market makers play in the creation of liquidity,
in this section we will explore what liquidity is, and some known ways of providing it.
Liquidity is the efficiency and ease with which an asset can be converted into another
one without affecting its price. In TradFi, the most liquid asset of all is fiat (cash), and
among them the most liquid is USD.
In traditional finance, liquidity is provided by large and sophisticated institutions, from
investment banks such as Goldman Sachs and Credit Suisse to High-Frequency Trading
(HFT) companies such as Citadel Securities, Jane Street, and Virtu Financial.
Those sophisticated algorithms and trading strategies1 competing against each other lead
to an extreme optimisation of the market making process. Strategies are devised by world-
1
A trading algorithm is a computer program that based on the data, such as price, order book,
external signals, risk-profile, execution costs, etc., decide what is the optimal portfolio to hold at any

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 10 of 88

class mathematicians and analysed for predictive capabilities. As for speed, specialised
hardware such as Field-Programmable Gate Array (FPGA) are programmed, hardware is
co-located in the exchange, the cable from the box to match making engine is measured
to be the same for each co-located participant, shortest possible paths to send signal
between exchanges are constructed, and much more. As a result, traditional markets
have liquidity provision optimized to the physical and computational limits.
In contrast, the DeFi scenario is quite the opposite. In TradFi all mathematical models,
hardware, and software are optimized to perfection, whereas in DeFi basic and simple for-
mulas and solutions are in use. This is purely due to how young and small (in comparison
to TradFi) the cryptocurrency market is, and especially DeFi.
The formula for Liquidity Provision in TradFi has geometric distribution, very well de-
fined higher moments, and is provided in very tight ranges. Contrarily, in DeFi most
liquidity formulas (such as CFMM) believe that it is as likely to buy asset X for $20, as it
is for $1, 000, 000, 000 or $0.000001. The inefficiency of the formulas utilized in DeFi are
the source of impermanent loss, capital inefficiency, inefficient and slow price discovery,
and loss to both market makers and market takers.

2.3 Efficient Market Hypothesis

Efficient Market Hypothesis (EMH) is a hypothesis stipulating that asset prices in the
market reflect all available information, or in other words, that the market is

• completely rational;

• that all participants have perfect access to information;

• that everyone’s beliefs can be reflected in the market due to the availability of
financial instruments (e.g., shorts).

However, this is often not true [122, 128, 123, 124, 125]:

• there is high asymmetry in access to information, from expensive research and

terminals to advanced analytical tools;

• there is limited access to reflect the market beliefs (e.g., shorting is only available
for sophisticated investors);

• agents do not act only on the information available, but also on their own personal
judgement, sentiment, fear, and many other emotions;

• not all agents are able to derive as efficient conclusion from the same data;

• agents are easily deceived by poor sources of information and market manipulation;

• inefficient and irrational methods are used to predict price movements;

given moment and rebalance to it. Trading strategy can be that algorithm or a rigorous model followed
by a trader such as in systematic trading, but can be as well non-systematic and rely much heavier on
the trader’s personal experience and reading of the market. TradFi market makers rely on sophisticated
and usually also very fast algorithms to facilitate market making.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 11 of 88

• investors are prone to take decisions hindered by different biases, e.g. loss aversion
in the case of trader who booked just a series of loses, confirmation bias in the case
of trader who just had a successful strike, overconfidence bias where trader believes
stronger in their abilities than the market data and signals, etc.

• and many more.

Taking into the account behavioral aspects of investing, 3 forms of EMH are proposed:
weak, semi-strong, and strong. In the case of weak form, the market price barely reflects
fundamentals and existing information. The cryptocurrency market specifically is in the
weak-form of EMH due to:

• small size when compared to the size of markets in TradFi where a single company,
Apple Inc, has larger market capitalisation than that of all crypto (on October 15,
2021), which amplifies the effects in the market;

• pricing models are flawed and lead to slower convergence on the market price, for
instance by the usage of CFMM[135];

• historical reasons, where early investors, called whales, accumulated disproportional

amount of wealth in comparison to other market participants, leading to a small
group of agents being able to significantly move the market;

• there is an evident lack of tools to reflect market sentiment, financial derivatives are
limited, and centralized exchanges (CEXs) are ineffectively implemented, leading
often to losses on non-losing shorts, etc.;

• cryptocurrencies are extremely volatile compared to other investments, this both

means that usually cryptocurrency traders are prone to accept higher level of risks,
but also that they are more exposed to the impacts of their own behavior on the
trading, as all the effects and biases are magnified[38, 39];

• there is a large network effect at play, where financial advice is derived from the
network effect, which is often exploited by malicious parties, who instead of focusing
on building fundamentally-sound products, rather focus on creating network effects;

• lack of clarity from the government on the regulation of crypto, preventing wider
adoption and causing investors to include other information than financial and
performance when making their predictions;

• inefficient taxation on cryptocurrency assets, which leads to agents often making

sub-optimal trades (e.g., sell a portion of good positions to secure fiat for future
taxes, minimize the amount of trades due to the accounting requirements, etc.);

• cryptocurrency markets are open 24/7, meaning there is an inevitable necessity for
active participants to exit the market or take sub-optimal positions in it, in order
to be able to sleep and fulfill their physiological needs – this is not an inane matter,
as active participation in the market requires sophisticated tools and is limited
by human capabilities (e.g., active crypto traders will often set up multiple alarm
clocks during the night to check on the key risk metrics and their positions);

• and many more.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 12 of 88

2.4 Automated Market Makers & Decentralized Exchanges

AMM is the underlying protocol powering all Decentralized exchanges (DEXs). AMM
is the autonomous market making mechanism which eliminates the need for centralized
exchanges, or large single parties taking the role of liquidity provider and creating all
related infrastructure themselves. AMM pulls the liquidity from the network and replaces
centralized market makers with code[134].
The unique property of a DEX is that nobody owns it per se, the protocol specifies how
providing liquidity and trading work. The code will be open sourced.
A vast majority of DEXs rely on CFMM to define the liquidity curve. Those are formulas
that specify how transactions are performed and how the price changes in response to
transactions taking place[133, 2, 3, 4].

2.5 Constant-Function Market Makers

CFMM was the first class of AMM applied to real-world financial markets. All CFMM
formulas have one distinguishing feature, that is they are all equal to a constant value,
for instance:

• Uniswap’s CFMM is x · y = const [5, 6]; more specifically (including slippage)

Uniswap’s formula is (Rα − ∆α )(Rβ + γ∆β ) = const = k; Uniswap’s formula is used
by multitude of other DEXs such as PancakeSwap;

• Bancor’s CFMM is an extension of Uniswap´s formula to any number Q of assets N all

sharing the same pool [4], also called constant mean market maker, N
i=1 xi = const;

• Hybrid CFMMs, which are modified in such a way as to achieve the desired prop-
erties based on the characteristics of the assets being traded.
n+1
Ann i xi + D = ADnn | + nnDQ xi , where x are the asset reserves, n is the number
P
i
of assets, D is an invariant representing the value in the reserve, and A is the
“amplification coefficient”, which is tuneable constant.
This produces an effect similar to leverage, and influences the range of asset price
volatility[39] (the higher the asset volatility the higher A should be set)[132].

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 13 of 88

Figure 1: Constant product AMM formula (Uniswap’s model).

Figure 2: Constant mean AMM formula (Balancer’s model).

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 14 of 88

Figure 3: Hybrid CFMM of Curve’s stableswap vs constant-product CFMM comparison.

2.6 Concentrated Liquidity

In unconstrained liquidity, i.e., one provided in the full CFMM range of (0, ∞) capital is
highly inefficient, for instance in Uniswap the trading pair DAI / USDC assigns only 0.5%
of the liquidity for the trading range between $0.99 and $1.01 through which majority of
trading volume goes through. This basically means that 99.5% of the liquidity sits idle.
Only about 0.5% of liquidity is enough to provide the same level of experience. What’s
more, that range of [$0.99, $1.01] is where market makers earn the majority of their fees.
Concentrated liquidity attempts to resolve this issue by allowing market makers to specify
the price ranges (indexes) between which they want to provide liquidity[7]. For instance,
in the case of DAI / USDC pair, they could specify the range to be $0.99 and $1.01, hence
concentrating the provided liquidity resulting in higher profit per unit of value locked.
Furthermore, such fragmentation of the pool naturally leads to emerging properties such
as geometric price distribution.

Figure 4: Convergence onto geometric price distribution of concentrated liquidity model.

Fragmentation leads to much more complex routing. In the case of Ethereum, where
all code execution happens on-chain, with a global memory (and hence high transaction
costs), this also has the inadvertent effect of increasing transaction prices. As the user
has to specify the price range when creating a concentrated liquidity pool, for weakly
correlated and non-correlated pairs the range will become obsolete quickly. This will

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 15 of 88

require the user to unstake their liquidity and resubmit it at the new range, spending
even more gas (and hence taking away from the market making profits they generated).
In contrast to these problems on Ethereum, there is fertile ground for concentrated liq-
uidity on Cardano. Due to the nature of the Extended Unspent Transaction Output Model
(EUTxO) model, concentrated liquidity can be smoothly implemented in such a way that
the off-chain code ensures the expensive computation can happen off-chain and only be
validated by on-chain validators, resulting in no impact on transaction costs from complex
routing.
That is exactly what Maladex achieves via its programmable swaps, defined in section 4,
and goes one step beyond with sophisticated and dynamic liquidity pool modelling, called
Algorithmic Automated Market Maker, as outlined in section 6.

2.7 Capital Efficiency ξ

Capital efficiency is a metric of how much capital is needed to generate the same level of
revenue. It is easier to think of money M as stored energy that can be used to perform
work, that work in turn is performed in expectation of generating revenue. Therefore,
the less energy E that’s required to perform the same work as W , the higher capital
efficiency ξ the system has.
For instance, in subsection 2.6, we described an example of providing liquidity to DAI
/ USDC pair. As the used CFMM formula provides liquidity over the range of (0, ∞),
but 3σ of trading activity takes place within the $0.99, $1.01 range, it means that only
around 0.5% of capital is used on useful work (providing liquidity in the active range),
thus 99.5% is unused except for the 1 − 3σ case (less than 1% of all the cases). Hence, the
capital efficiency of providing liquidity for the DAI / USDC pair using x ∗ y = k formula
is ξ ≤ 0.5%. Imagine a car engine that only uses 0.5% to propel the car and loses 99.5%
of fuel in unproductive energy, such as heating the engine.
This issue is prevalent across all CFMM AMM and makes Total Value Locked (TVL) un-
productive. One of the main premises of the Maladex protocol design with programmable
swaps and Algorithmic Automated Market Making (AAMM), is to put capital to efficient
use, the same way as TradFi market makers do.

2.8 Impermanent Loss

Impermanent loss is an integral part of all CFMM AMM and is rooted in the const = k
part of the equation, which entails that ratio of the assets in the pool changes with the
asset price, and there is no mechanism to counter it. This means that for the majority
of the trading pairs, besides special cases such as pools of all assets following the price
of the same external asset (underlying) e.g., same currency stable coins, like DAI, BUSD,
USDC, USDT, etc., the impermanent loss is guaranteed.
We cover impermanent loss in great detail in subsection 6.1. However, here we will just
outline what causes it and what effect it has on the capital locked into an AMM protocol.
Why are we so concerned with impermanent loss? Market makers earn money by pro-
viding liquidity, usually they earn a portion of the protocol fee, equivalent to 0.1 − 0.2%
of the exchanged assets. This provides compensation for providing liquidity. However,

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 16 of 88

users always have the alternative of keeping their assets in their wallets instead of locking
them in a liquidity pool. Now, as the asset ratios in the pools change, impermanent loss
starts to occur, as summarized in Figure 5.

Figure 5: Percentage Divergence Impermanent Loss (Duplicate of Figure 17 for Easier

Reading).

The main challenge is the exponential nature of impermanent loss. For small asset ratio
variations, the loss is negligible, but for large shifts it easily eradicates all the profits from
providing liquidity and from governance tokens earned from yield farming.
Maladex aims to eliminate impermanent loss completely (property derived from the pro-
tocol design), and where the user desires to provide liquidity via a specific model (even
including CFMM), it provides means to statistically eliminate it as well via asset ratio
management, arbitrage, and dynamic risk compensation, all outlined in subsection 6.1.

2.9 Risk Control

Risk is inseparable from any investment, but risk control is not part of any existing AMM
architecture. Risk control activities include:

• portfolio allocation that achieves the optimal expected return given fixed maxi-
mum accepted level of risk σ such as done in Markowitz’s Modern Portfolio Theory
(MPT)[27, 26];

• providing tools and data to the user when engaging in the market, e.g., creating
liquidity pool, helping to achieve a much better outcome;

• actively calculating and monitoring risk metrics such as VaR (Value at Risk), volatil-
ity, Sharpe ratio, etc.[40, 41, 43];

• dynamically hedging the risk, e.g., buying inversely correlated assets to offset the
investment risk, and updating this hedge dynamically to maintain the risk level
below desired levels;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 17 of 88

• providing risk-taker compensation, hence helping investors (e.g. market makers),

to maintain risk neutral positions;

• integrate fuses and other market stress scenario control mechanisms, including off-
setting sudden crashes to other parts of the network (e.g. when a rug-pull is iden-
tified, performing swaps on all other DEXs to convert and remove the risky asset
from the portfolio[27]);

• provide verification of policy ids, but also trust scores, and ability to whistleblow.

We delve deeper into how the Maladex platform enables management of risk in section 14.

2.10 Cardano Blockchain & EUTxO Model

Cardano offers 3 major innovations, making many of the ambitious goals outlined in this
paper achievable, which were previously impossible on global-shared state blockchains:

• Smart contracts are composed of on-chain validators and off-chain code; on-chain
validators provide the same level of assurance as any other blockchain model, but
at the same time Turing-complete off-chain code, enables performing complex and
resource-consuming computation without any impact on transaction cost; this is
revolutionary, as ideas such as programmable swaps and algorithmic automated mar-
ket maker (AAMM) would not be possible to implement without it. What’s more,
on-chain validators provide full scope of smart contract security, meaning that de-
spite the off-chain code, the entire protocol is equivalent to being fully executed
on-chain as all security is achieved on-chain.

• EUTxO fragmentation and redeemer model, which provides a unique method to

enable many concurrent independent actions to take place in parallel across multiple
protocols; for instance, given this fragmentation and desire to arbitrage with other
DEXs, this can be done simply thanks to the EUTxO model and how redeemers
work.

• Hydra head – fast and isomorphic state channels allowing for straightforward usage
of the same transaction settlement code in the local hydra head formation. This
enables a significant performance improvement without incurring layer 2 software
engineering cost.

2.11 What’s Next

After having inspected the current state of AMMs and having outlined existing challenges,
in the following sections we discuss the Maladex protocol implementation objectives that
aim to address the aforementioned challenges.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 18 of 88

3 Maladex Protocol Mission & Objectives

Maladex aims to revolutionize the world of finance by:

• providing the tools that allow everyone to create sound and efficient ways of invest-
ing;

• increasing the efficiency of the cryptocurrency market and accelerating the transi-
tion from TradFi to DeFi;

• building the foundations of the financial markets of the future.

3.1 Providing the Building Blocks of Sound and Efficient Ways

of Investing to Everyone
Maladex will be a sound platform for developing investments with high capital efficiency
and without impermanent loss. Additionally, Decisions and observations will be sup-
ported by data, market indicators and by providing a DeFi education platform.
We aim to develop a pathway for people to improve their investing skills, as well as
providing tools and infrastructure for the creation of revenue such as market making,
crypto indexes, synthetics, automated portfolio management strategies, and more.

3.2 Increase Market Efficiency

The Maladex platform improves market efficiency by allowing users to create a range
of products, allowing for the expression of specific market participant sentiments, from
swaps to option strategies and sophisticated automated trading strategies.
Markets that cannot be shorted are inadvertently inefficient [22, 18, 19]. This is a simple
consequence of the fact that a person who believes that asset X is undervalued can buy
it and hold it until the price discovery catches on, while the person convinced that the
asset is overpriced cannot take a separate position (short). The person who believes asset
X to be overpriced cannot simply sell it, as it only protects them from the downside risk,
but does not provide a way to capitalize on their knowledge of the asset overvaluation.
This results in very little incentive for discovering overpriced assets, but there’s one for
undervalued ones. If you can only add water to a jug, but never take away, the amount
of water will inadvertently only increase. It can be even stipulated that a lot of volatility
in the crypto market takes origin in this mechanic. Due to a lack of sufficient access to
short instruments, the price tends to go up until it suddenly reverts to the mean, like
stretching a rubber band to its limits[39].
Developing a sound system composed of all types of financial instruments, most impor-
tantly negatively correlated ones, will lead to increased market efficiency and drive it
closer towards the equilibrium.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 19 of 88

3.3 Build the Foundations for the Future Financial Markets

Last, but not least, having a trading platform with a wide range of tradeable crypto assets,
indexes, derivatives, and mirrored assets, enable the development of more specialized
financial platforms on top of this infrastructure layer. One example of such a layer
would be the creation of a distributed hedge fund, allowing users to stake investments in
automatic trading strategies and bots, providing exposure to a wide range of additional
source of alpha (additional revenue compared to an index without taking additional risk).
We start our discussion of novel proposed ideas with programmable swaps – a combination
of off-chain and on-chain code capable of performing many functionalities, from market
and limit orders to executing sophisticated automated trading strategies.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 20 of 88

4 Programmable Swaps
In this section, we introduce the concept of programmable swaps, swaps composed of:

• type - order type;

• triggers - specifying conditions upon which the swap is active to be executed;

• actions - payload definition of what action a programmable swap performs upon

activation;

• assets - assets required to perform actions outlined above.

Programmable swaps provide an elegant and automated way for the execution of many
order types, just to give a few examples:

• market order (a.k.a. swap) – an order that is executed immediately, exchanging

one asset for another, e.g., $2 for 1 ADA; this programmable swap has no triggers
hence it is always active; it has one action, exchange one asset for the other at the
best immediately available market rate, and locks assets specified for the exchange.

• limit order – an order that executed at specified price or better; limit order can
be executed partially; this programmable swap has 1 trigger - the best available
price is specified by least upper bound (LUB) for sell orders (the lowest price at
which a trader is willing to sell the asset) and greatest lower bound (GLB) for buy
orders, the maximum price for which a buyer is willing to buy a specific asset; this
programmable swap is composed of order type limit-order, LUB /GLB (for sell, buy
respectively), and assets to swap;

• weighted DCA (dollar-cost average) is a trading strategy that exchanges at defined

intervals, one asset for the other, adjusting the daily portion of exchanged assets
(weight) for given market conditions (e.g., when the price rises buy less and when
the price drops buy more); this programmable swap is composed of a time trigger
defining all time indexes after which the DCA operation should be performed (e.g.,
daily interval). Actions are composed of market indicator Oracles (e.g., divergence
and momentum indicators indicating local price minima and maxima), and DCA
parameters (periods, the response to indicators, etc.); finally it contains USD or
other assets, which we cost average into the other asset (e.g., ADA).

• engage arbitrage bots that exploit pricing inefficiencies of models used by other
exchanges, to both increase the profits of arbitrage bot funders, increase the overall
market efficiency and to ensure that the Maladex platform reflects the true market
price, regardless of TVL (Total Value Locked);

• dynamic liquidity provision - is risk-controlled market making where in stable con-

ditions and price ranges (medium price and volatility) assets are provided bidirec-
tionally (buying and selling at the same time, i.e., taking both positions in the
market around the current price and in the given spread around it), providing liq-
uidity to the market and earning the market maker fees. What is unique about
this approach is that as soon as the price of one asset (e.g., ADA) starts to quickly

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 21 of 88

rise, the programmable swap is converted into ADA (the asset that investor prefers
to hold) and the swap is terminated. This allows the market maker the ability
to earn rewards from market making, but prevents being locked out of the pre-
ferred asset as its relative price to USD starts to rise. This programmable swap is
composed of a trigger that defines the safe trading range, once it is crossed, the
programmable swap isn’t executed anymore, actions – provide liquidity according
to efficient market formula, all assets being exchanged;

• portfolio management strategy – a complex trading strategy defining system for

performing trades, taking the market data as input and outputting the actions to
rebalance the portfolio[27, 26]. This type of strategy usually employed by sophisti-
cated traders. It allows for the implementation of systematic investment strategies.
Such strategies are commonly employed by traditional market makers and hedge
funds. This type of programmable swap is composed of triggers representing fuses,
actions consuming a wide array of Oracle data, describing all inputs necessary for
the strategy to design portfolio rebalancing decisions, the strategy definition trans-
forming Oracle data into new portfolio balance, initial assets (e.g., ADA) to seed the
trade, and a current portfolio allocation thereafter. It is worth mentioning that this
type of programmable swap allows for the implementation of hedge funds on-chain
by creating a market for portfolio managers (PMs) to develop trading strategies,
and for investors to provide funds via locking them into trading strategy vaults,
and paying a commission to PMs from the generated profits; this creates a unique
opportunity for everyone having access to services of seasoned PMs and for users
to have access to the best performing strategies in the market.

• dynamically hedge your position using derivatives/underlying/etc..

4.1 Architecture
The Maladex programmable swap protocol is composed of 2 parts:

• the commit phase – send commitment onto the blockchain to perform a certain
action; each commitment can be cancelled by sending a cancel order as another
programmable swap; each commitment is an NFT and can be produced in parallel
using a minting policy;

• the execution phase – all commitments are executed against each other each block,
resulting in trades happening; due to high fragmentation and optimized routing,
transactions take much less memory and can be efficiently managed.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 22 of 88

Figure 6: Maladex programmable swap protocol execution stages.

4.2 Commit Phase

All programmable swaps are submitted onto the blockchain by minting NFTs represent-
ing the programmable swap and all internal information. Minting is an action that is
perfectly parallelizable (see *parallelism) – NFTs are picked from the user´s wallet to
mint commitment and as many NFTs as the memory pool allows. These can be minted
in parallel each block cycle.
Now, all commitments represent intents to perform certain actions given specific market
conditions (triggers). All commitments are separated into 2 types: - active frontier –
programmable swaps with all triggers active, hence executable this block cycle; - inactive
frontier – programmable swaps with at least one trigger not-active, hence not executable
this block cycle.
As Maladex programmable swap protocol uses hydra heads to scale the number of trans-
actions, it is worth describing the consequences of this beautifully simple, fragmented
design. All EUTxOs present in active frontier can be moved to the hydra head for exe-

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 23 of 88

cution and once executed, back to layer 1 to settle the trades on the base layer. As each
programmable swap is the smallest possible intent, it presents the most optimal exchange
of information between layers.
What’s more, this design means that even when executed on layer 1, much less memory
needs to go into all trades. The minimum possible amount of information is expanded
in the execution of programmable swaps, hence the memory used and cost is minimal as
well. Traditional AMM models, with a single EUTxO representing the entire liquidity
for a given pair, need to be included in all transactions, leading to large transaction sizes,
and hence larger costs and lower total throughput (or TPS - transactions per second).

4.3 Execution Phase

After commits are submitted onto the blockchain in the form of Non-Fungible Tokens
(NFTs), they enter the execution pool. The execution pool is composed of 2 disjoint
groups:

• active frontier A - the set of all programmable swaps schedulable in the current
block, defined as A - the set of all programmable swaps of which all triggers are all
active, defined as A(b) := {s ∈ S | kactive triggers(s, b)k = ktriggers(s, b)k} = {s ∈
S | kinactive triggers(s, b)k = 0};

• inactive frontier I - the set of all programmable swaps of which at least one trigger
is inactive, defined as I(b) := {s ∈ S | kactive triggers(s, b)k < ktriggers(s, b)k} =
{s ∈ S | kinactive triggers(s, b)k > 0}

where

• b - block id;

• s - programmable swap;

• S := A ∪ I - set of all committed programmable swaps;

• active triggers(s, b), inactive triggers(s, b), triggers(s, b) - functions that return cor-
respondingly all active, inactive, and just all the triggers for the given swap s during
the block b.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 24 of 88

Figure 7: Active and inactive frontiers.

The Cardano blockchain architecture specifies the smart contract architecture of:

• on-chain validators – scripts validating that the submitted transaction, with speci-
fied input and output EUTxOs, can be executed; this part is executed on-chain by
block producing nodes;
• off-chain code – Decentralized Application (dApp)’s code, responsible for all plat-
form functionality and most importantly, preparation of the transaction to be sub-
mitted onto the blockchain, to be validated and executed by validators, and con-
firming the desired state transition.

Cardano’s blockchain design makes it very straightforward to implement complex off-

chain logic, in the case of Maladex it is this order matching engine responsible for exe-
cuting all programmable swaps according to their specification and the on-chain state.
Such off-chain component requires:

• Plutus Application Backend (PAB) (or its equivalent) to interact with Cardano
nodes and a smart contract enabled wallet via dApp connectors;
• smart contract enabled wallet (Cardano wallet capable of storing dApp endpoints
and interacting in the smart contract execution);
• Cardano node to monitor and query the blockchain.

From the point of view of Maladex protocol implementers, we see this architecture as
highly beneficial, as it allows for the implementation of an advanced order matching
engine, which we outline in the below section, subsection 4.4, and expand in full detail
in section 5.

4.4 Maladex Execution Pipeline

Maladex execution engine takes a look at the latest snapshot of the blockchain layer 1
and then in order:

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 25 of 88

• processes cancel order requests and cancels all non-EOL (end of life) programmable
swaps;

• partitions the set of all programmable swaps S into the active frontier A and inactive
frontier I;

Figure 8: Maladex protocol (off-chain order matching engine) execution pipeline.

4.5 Scalability
The process of minting NFTs can be done completely in parallel (see *parallelism). Hence,
it is possible to create a programmable swap, and interact with the Maladex protocol, with
minimal risk of delays and requirement for sequencing. NFT minting is parallelizable.
Once minted, NFTs storing the programmable swap code and assets required to transact
are ready to interact with the market. Hence, the commitment phase achieves the highest
level of concurrency possible (full parallelism).
On the other hand, in the execution phase, multiple EUTxOs can be matched against
many other EUTxOs, meaning that we need to implement an efficient routing mechanism.
This is explained in detail in section 5. Due to optimal EUTxO fragmentation and graph-
based execution scheduling, it results in the most scalable routing mechanism possible
(limit of which is defined by Amdahl’s law).
The proposed scaling design is not only an elegant solution to the concurrency challenge,
stemming from the lack of global memory (distributed state) and the ability to spend only
once each EUTxO per block, but also one resulting in a much smaller memory footprint
as only the information required for the execution of the order goes into the transaction.
This is the opposite of the popular Uniswap-style designs, which in the context of the

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 26 of 88

EUTxO model, leads to including all available information for the given liquidity pool
into a transaction (which is the least memory-efficient solution possible).

4.6 Memory Requirements

The Cardano protocol requires a payment of a · size(tx) + b where a is proportional-
ity constant representing the cost of a unit of storage and b is a minimum payable fee
(introduced with the primary goal of preventing Distributed Denial of Service (Attack)
(DDoS) attacks)[129]. Therefore, it is important and beneficial to all Maladex protocol
participants for all transactions sizes to be minimal.
In centralized liquidity models (the ones relying on single EUTxOs), the total amount
of memory required for the transaction is high. All of that centralized information is
included in every transaction that interacts with the EUTxO that stores the liquidity
pool information.
In contrast, the Maladex protocol, does quite the opposite. Instead of including all exist-
ing information as centralized EUTxO liquidity models do, only the bare minimum re-
quired information is included. The Maladex protocol is composed of highly fragmented
programmable swaps and as those are matched with each other, only the information
required to perform the action is stored in all. The consequence of this is much smaller
transaction size requirements.

4.7 EOL of Programmable Swaps

Once all tasks outlined in the programmable swaps are done, it reaches its EOL – end of
life. This means that the programmable swap can be destroyed and is no longer needed.
There are two conditions for termination of programmable swaps: cancel order (order
to cancel existing programmable swap) and end of life (EOL) (when the programmable
swap has completed its task).
When this happens, a final transaction is performed where:

• all assets in the programmable swap are returned to the wallet from which they
originated;

• a receipt NFT is minted and sent alongside.

4.7.1 NFT Receipt

NFT receipt is an NFT containing the information about how the trade was executed
and might contain a graphical visualisation of the performed tasks, for instance:

• market orders (swaps) and limit orders will just include information about the
execution price and time;

• Dollar-Cost Averaging (DCA) would contain a record of all buy orders, assuming
DCA is executed daily over a period of 30 days, that would contain 30 records
similar to market order / limit order outlined above;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 27 of 88

• liquidity pool fragment minting using one of the available formulas and parametri-
sation (stochastic models) will contain the pictorial representation of the model
(distribution, its moments, and sensitivity to market dynamics), etc.

NFT receipts will fulfill 2 important roles:

• book-keeping – providing information to the user about how the orders were exe-
cuted;

• Profit & Loss (P&L) tracking – tracking performance of the executed strategies;

• collectibles – best trades (achievements), first trade to exchange some newly listed
asset, etc.

4.8 Theoretical Model Performance Characteristics

The proposed, commit-phase solution maximises concurrency and provides means to
achieve high throughput. This is parallelization, commit-phase submissions are performed
in parallel, subsection 4.2, and the optimal fragmentation of EUTxO information in the
execution phase, subsection 4.3.
The proposed solution allows for full parallelism during minting commit-phase NFTs
of programmable swaps. This in the result achieves the highest score using Amdahl’s
law metric. This further leads to improved ability to parallelise transations and reduce
memory footprint on both layer 1 and 2. The burden of work is off-loaded to order
matchmaking engine, which by doing so, is able to implement Domain-Specific Language
(DSL) for programmable swaps, and provide a novel utility to the universe of DEXs.

4.9 Fragments Matching

The automated and algorithmic execution offered by programmable swaps is fully executed
on-chain. There are no links to external APIs, there is no external software that has access
to the contracts, there is no critical component stored off-chain. It is worth noting that in
Cardano’s on-chain validator and off-chain code programming model, while the off-chain
component is responsible for selecting EUTxOs for the transaction, the actual execution
still happens on-chain. This means that the model offered by the Maladex protocol is not
only self-contained, but more importantly, it is trustless and implemented using smart
contracts.
In this section, we explore further how the fragments are matched against each other and
how fairness and security are achieved.
We are going to explore fragmented liquidity pools, without explanation of what Algo-
rithmic Automated Market Making (AAMM) is or how it works, which we will explain in
detail in section 6. Here, we use the same model as in section 6, namely the fragmented
liquidity model, but focus on the execution, trustlessness, and safety.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 28 of 88

Figure 9: Fragmented (virtual) liquidity pool’s initial state S0 .

We start with fragmented liquidity pool2 initial state S0 .

Each bubble above is a separate EUTxO, as defined by an NFT minted in the commit
phase. In the example below we used limit orders (circles) and AAMM fragments (AMM
liquidity pools automatically and algorithmically adjusting to the market conditions).
We’ll discuss 3 execution examples:

• market order matched by limit order;

• market order matched by AAMM liquidity pool;

• limit order matched by limit order on the opposite side of the order book.

4.9.1 Market Order Matched by Limit Order

Let’s assume market order M0 being committed in the previous block B−1 .The Market
order is executed as long as there is liquidity, at the best available price. In this example,
let’s assume that the best available price is provided by limit order LS,1 and that the
amount of assets being sold by LS,1 is sufficient to cover the entire M0 (if it wouldn’t
2
Fragmented liquidity pool can also be called virtual due to the fact that it is not a pool, but multiple
fragments (separate EUTxOs) from which properties a liquidity pool naturally emerges (by liquidity pool
we mean here pooled resources that give the impression of the market being liquid.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 29 of 88

simply the next best available limit order or AAMM would be used, whichever provides
the best price).
Now, a transaction T0 is submitted that uses both M0 and LS,1 as input EUTxOs and
creates as output:

• EUTxO with swapped assets, spendable by the agent that submitted M0 (hence
swapped assets are directly sent to that user´s wallet);
• a new EUTxO with remaining assets from LS,1 after executing M0 .

Figure 10: Transaction executing M0 using LS,1 .

After executing T0 the state of the fragmented liquidity pool is as below:

Figure 11: State of the fragmented liquidity pool after executing T0 .

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 30 of 88

As we can see LS,1 and M0 are spent, LS,E with remaining liquidity is created and * the
result of market order (swap) is returned to the wallet of the M0 submitter3
The presented approach has multiple benefits:

• thanks to commit phase the transaction M0 is submitted without any delays and
bereft of concurrency issues immediately onto the ledger;

• off-chain components match LS,1 with M0 , which provides the best execution price;

• the transaction created by off-chain match making engine has minimum possible
size, only EUTxO with the market order (swap) is consumed and only single EUTxO
providing liquidity. In contrast to entire liquidity pools stored in a single EUTxO
(the biggest possible footprint of transaction memory per transaction) it offers the
highest possible optimisation. In a sense, this is not much different from a regular
transaction on the Cardano blockchain, which automatically allows high scalability
of Maladex on layer 1; this cost memory size minimisation means that market taker
pays the smallest possible transaction fee for having their transaction executed;

• limit order is free of impermanent loss and represents the price the user wants to
get for the exchanged assets, plus provides market maker fees to the limit order
owner;

• programmable swaps allow for additional parameters to be specified in a limit order,

e.g., to be executed at S + 1 price, as opposed to fixed price, this price is moving
with the market price, offering constant market making opportunities until resource
exhaustion.

From the security point of view:

• sell limit order LS,1 is only spendable at the preset price of S + 1 in this example,
hence there is no other way to redeem this EUTxO than at this price;

• market order M0 is guaranteed to be executed at the best market price by the

transaction input of the transaction at the current market price plus/minus accepted
volatility: S ∓ σ.

4.9.2 Market Order Matched by AAMM -LP

Let’s assume market order M1 being committed in the previous block B−1 . Market order
is executed as long as there is any liquidity, at the best available price, same as in the
previous example, subsubsection 4.9.1, however now the best available price is provided
by Algorithmic Automated Market Maker (AAMM) A3 .
Now, a transaction T1 is submitted that uses both M1 and L0 as input EUTxOs and
creates as output:

• EUTxO with swapped assets, spendable by the agent that submitted M1 (hence
swapped assets are directly sent to that user´s wallet);
3
The assets are not sent per se, simply the output is only spendable by the creator of M0 wallet
address, which is the EUTxO-native way of returning assets.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 31 of 88

• a new EUTxO with updated assets and distribution A5 .

Figure 12: Transaction executing M1 using A3 .

After executing T0 , the state of the fragmented liquidity pool is as below:

Figure 13: State of the fragmented liquidity pool after executing T1 .

We can see that A3 and M1 are spent, A5 with updated asset balance and distribution
model has been created and * the result of market order (swap) is returned to the wallet
of the M1 submitter.
The presented approach has multiple benefits:

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 32 of 88

• thanks to commit phase, the transaction M1 is submitted without any delays and no
concurrency issues, immediately onto the ledger (this is the same as for the previous
example, and is an invariant that holds regardless of the transaction type);

• off-chain components match A3 with M1 which provides the best execution price;

• the transaction created by the off-chain match making engine has the minimum
possible size, only the EUTxO with the market order (swap) is consumed and
with only a single EUTxO providing liquidity. In contrast to entire liquidity pools
stored in a single EUTxO (the biggest possible footprint of transaction memory
per transaction), it offers the highest possible optimisation, this is the same as a
regular transaction on the Cardano blockchain. which allows high scalability of the
Maladex Protocol on layer 1; this cost memory size minimisation means that the
market taker pays the smallest possible transaction fee for having their transaction
executed;

• Algorithmic Automated Market Maker (AAMM) is a unique model that collapses

the provided range to the true geometric price distribution of the asset, hence is
free of impermanent loss and represents the price that the user wants to get for the
exchanged assets, plus provides market maker fees to the AAMM liquidity fragment
owner4 ;

• programmable swaps and liquidity pool moment-based distribution allows for smooth
adjusting based on market conditions and the pool composition, providing contin-
uous liquidity at the true market price plus minus volatility S ∓ σ, for all requests
coming at the prices close enough to the true market price, or better.

From the security point of view:

• sell limit order A3 is only spendable in the present geometric distribution around
mean S and within the volatility σ, further precised by the higher moments of
the distribution (skew, kurtosis, etc.), hence there is no other way to redeem this
EUTxO than at the fair and efficient market price plus/minus market volatility σ;

• market order M1 is executed at the best market price by the transaction input of
the transaction at the current market price plus/minus accepted volatility: S ∓ σ.

4.9.3 Limit Order Matched by the Order Book

The final example deals with limit orders matching each other, this happens when the
order book sides cross over (that is if there is a buy order with the price equal or greater
than the cheapest sell order), i.e.

∃b∈Ib ,s∈Is Lb .price ≥ Ls .price

4
As identified by the wallet which minted the liquidity fragment.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 33 of 88

Figure 14: Order book crossover.

In the figure provided, the following 3 limit orders cross over: LB,1 .price = LB,2 .price =
LS,1 .price. That means they should execute immediately among each other. They
provide liquidity to each other, and in this case the total volume of LS,1 .volume >
LB,1 .volume + LB,2 .volume meaning that both LB,1 and LB,2 will be executed completely
and the remaining liquidity of LS,1 will return as a fragment to the virtual liquidity pool.
Now, a transaction T2 is submitted that uses LB,1 , LB,2 , and LS,1 as input EUTxOs and
creates as output:

• 3 EUTxOs with swapped assets, spendable by agents that submitted by LB,1 , LB,2 ,
and LS,1 respectively (hence swapped assets are directly sent to that user´s wallet);

• a new EUTxO with remaining assets from LS,1 after executing LB,1 and LB,2 .

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 34 of 88

Figure 15: Transaction matching 3 order book transactions: LB,1 , LB,2 , and LS,1 .

After executing T2 the state of the fragmented liquidity pool is as below:

Figure 16: State of the fragmented liquidity pool after executing T2 .

We can see that all LB,1 , LB,2 , and LS,1 are all spent, LS,E with remaining liquidity is
created and * the results of limit orders are returned to the corresponding wallet owners.
The presented approach has multiple benefits:

• as before, thanks to the commit phase design, all LB,1 , LB,2 , and LS,1 could be sub-
mitted at block B−1 and executed at block B0 right away, without any concurrency
and memory related issues;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 35 of 88

• off-chain components match LB,1 , LB,2 , and LS,1 together, which provides the best
execution price for all 3 orders;

• the transaction created by the off-chain match making engine has the minimum
possible size, only EUTxOs with the required assets and data are consumed, with
only a single EUTxO providing liquidity. In contrast to entire liquidity pools stored
in a single EUTxO (the biggest possible footprint of transaction memory per trans-
action), it offers the highest possible optimisation. In a sense, this is not much dif-
ferent from a regular transaction on the Cardano blockchain, which automatically
allows high scalability of Maladex on layer 1; this cost memory size minimisation
means that market taker pays the smallest possible transaction fee for having their
transaction executed;

• limit order is free of impermanent loss and represents the price the user wants to
get for the exchanged assets, plus provides market maker fees to the limit order
owner;

• programmable swaps allow for additional parameters in a limit order to be specified,

e.g., to be executed at S + 1 price, as opposed to a fixed price, this price is moving
with the market price, offering constant market making opportunities until resource
exhaustion.

From the security point of view, sell limit order LS,1 is only spendable in the present price
of S + 1 in this example, hence there is no other way to redeem this EUTxO than at the
set price.

4.10 Programmable Swaps Domain-Specific Language

Programmable swaps on the Maladex platform will be based on DSL, unless precluded oth-
erwise by optimisation. This language will allow users to interface with various protocols
on the Maladex platform. This language will be writting in the spirit of the functional
programming pearl written by Simon Peyton Jones et al, “Composing Contracts: An
Adventure in Financial Engineering” [1].
The idea behind DSL is 3-fold:

• develop a secure and easy to use DSL for defining all potential trading strategies;

• provide means for composing financial contracts, creating a sound financial system
(you compose contracts, hence there is no material risk as in TradFi);

• be able to develop an easy GUI (Graphical User Interface) on top of it, for the most
popular types of contracts, such as market orders (swaps), limit orders, DCA, etc.

Initially Maladex programmable swap programming language will be simple, to provide

the basic functionality on the launch date. Over time it will evolve into a system that
allows Portfolio Managers (PMs) to define sophisticated automated trading strategies,
akin to hedge funds.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 36 of 88

4.11 User Interface

The programmable swaps DSL is an advanced feature, making all programmable swaps
features available to the users, however it is of a much higher level of complexity than
should be expected for a typical user (the biggest target group).
Maladex platform will to implement interfaces allowing users to easily specify desired
programmable swaps, without the need to understand programmable swaps DSL. Hence
Maladex will have:

• a simple market order (swap) interface for selecting the trading pair and specifying
the exchanged amount;

• a simple limit order interface for specifying the price bounds and the amounts of
tokens to transact;

• a simple interface for DCA specifying the interval, frequency, and methodology,
etc.;

• an option strategy visualisation graph, defining the underlying programmable swap

by moving points on the graphical representation, see subsubsection 11.3.5 for ex-
ample illustrations, in those cases, strikes K, expiry date, etc. would be movable
parameters and displayed in the generated graph.

What is more, based off of programmable swaps, DSL will allow for specification of end-
points, and thus automated generation of user interfaces.

4.12 Fundamental Building Blocks

A unique property of programmable swaps is their composability. Composability of fi-
nancial instruments ensures a sound definition [1], protected from over leveraged and
unbacked structures that often lead to economic ruin. However, programmable swaps
composability means that they can be combined with each other, type checked, and
result in new unique programmable swaps.
This fundamental design allows for all Maladex trading features to be implemented using
programmable swaps, for instance:

• an index as a programmable swap, that includes the rules for index composition
(what assets and in what amount go into the index during each rebalancing opera-
tion), and the assets that the index is composed of;

• synthetics, which might include just the rules for the asset collateralization, liqui-
dation triggers, and the collateral;

• options and futures, which contain assets in question locked into the contract, au-
tomatically settled upon the option expiration;

• option structures that can contain just different options defined at different strikes
and expires, and code that defines their composition;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 37 of 88

• sophisticated portfolio management strategy, akin to those used by hedge funds,

containing the assets locked for the strategy via a vault, and rebalancing rules;

• and many more, the design scales to any types of financial instruments.

Composability is the key element of functional programming and composing programmable

swaps is the most functional programming way of implementing financial contracts. How-
ever, what is most important is the safety that comes from both the compilation process
and inclusion of dependencies in the programmable swap itself, as well as the scalability
of the work, where all previous work directly contributes to features built in the future.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 38 of 88

5 Maladex Order Matching Engine

The Maladex protocol’s Order Matching Engine is the off-chain component responsible
for matching programmable swaps with each other according to their specification. In
this section, we cover it in more detail.

5.1 Architecture
Maladex Order Matching Engine is an independent component deployed on machine with
Cardano Node. Cardano Node provides the state and ability to query the blockchain,
the functionality required to get the list of all existing programmable swaps S and to
communicate the orders (state transition).

5.2 Routing Rules and Fairness

The engine deals with multiple types of orders such as:

• market order – execute at the immediately available price, given that within the
slippage range;
• limit order – execute only if at given or better price;
• liquidity pool – providing liquidity to the market at the current price;
• DCA – slightly more sophisticated order types, in this case market order executed
T
f
(where T is period and f is frequency) times over the span of the contract;
• sophisticated portfolio management strategy – composed of many conditions and
rebalancing strategies.

It is clear that due to the composability of the protocol, all that is required to guarantee
fairness of all programmable swaps is to guarantee fairness of the fundamental building
blocks, e.g., DCA will be guaranteed to be fair, if its’ building blocks, i.e., market orders
are guaranteed to be fair. This means that in the version 1 of programmable swaps we
only need to guarantee fairness of the the fundamental building block, market orders and
limit orders for all programmable swaps, regardless of how complex, to be fair.
The routing order to ensure fairness is as follows:

1. All cancel order requests are processed. This guarantees that if someone wishes to
cancel their order it is done before orders are executed. This creates a potential
vulnerability where a user might submit an order in one epoch and immediately
cancel it in the next, but a fee (denominated in MAL) for order cancellation will
remediate it.
2. All market orders are processed in the order of submission (FIFO - first in, first
out). For market orders to be processed the price must be within the requested
slippage.

The actions of more complex programmable swaps boil down to the guarantee of market
orders and limit orders, thus ensuring that all programmable swaps are fair.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 39 of 88

5.3 Security
The safety of programmable swaps is guaranteed by the conditions encoded in the rede-
meer, i.e.,

• order can only be cancelled from the wallet that created it (potentially in the future,
from a wallet that contains the appropriate $handle);

• order can only be executed if there is on-chain proof that it is the best available
price and within defined limits of execution, such as slippage.

Those metrics do not depend on any additional sources of security, but rather simply on
the market conditions. There is no more security needed, as all required conditions are
to be included within the redeemer and are provided already by Cardano blockchain.

5.4 Scalability
A naı̈ve implementation of the order matching engine would allow for deployment of
multiple instances N , all executing and submitting transactions in parallel. This would
lead to unnecessary transactions spamming the mempool with requests, and inefficient use
of resources. Only one transaction would be validated successfully onto the blockchain,
very inefficient usage of the Cardano infrastructure.
A proposal will be created to implement a consensus mechanism between the engine nodes
that elect the block leader (block leader as opposed to Cardano’s slot leaders), i.e., the
node responsible for submitting transactions in the next block.
A series of metrics for all engine nodes is going to be tracked, such as successful sub-
mission of all orders into the next block and in fair order, and upon any deviations the
underperforming node would be automatically deprioritised from the network, and in
extreme cases (e.g., submitting unfair ordering) permanently banned from the consensus
circle.
The ability to disconnect nodes is going to be guaranteed by the challenge included in the
nodes, which the disconnected node would not be able to acquire information required
to solve.

5.5 Decentralization
The aforementioned mechanism presents a unique ability to decentralize order execution,
by sharing the engine binary with Stake Pool Operations (SPOs). SPOs would then be
able to partake in order execution and hydra head (layer 2) formation for participation
in the DEX rewards.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 40 of 88

6 Algorithmic Automated Market Making (AAMM)

Algorithmic Automated Market Making (AAMM) is a novel AMM which emerges as a
property of the Maladex protocol in order to

• completely remove or strongly limit impermanent loss;

• increase capital efficiency by utilising the capital more efficiently and keeping it
constantly at work;

• use realistic pricing and liquidity curve models, as known from QuantFi and TradFi;

• dynamically adjust to market conditions and maintain liquidity pool properties via
dynamically adjusting its parameters;

• aid users in creating healthy liquidity pools and pairs;

• provide better means of price discovery;

• provide means of risk control.

Most established liquidity models do not model markets well (see model, EMH, and
impermanent loss in the glossary for a detailed rundown), leading to low capital efficiency,
expenditure of large amount of energy (see money ≡ energy) compared to the total value
locked, and have inherent scaling limitations due to global memory limitation (akin to
Ethereum´s account-based model architecture). The AAMM model attempts to bridge
that gap by applying more financial engineering scrutiny and bespoke methodologies.

6.1 Impermanent Loss & Constant-Function Market Maker

Impermanent loss originates from the inherent inefficiency of AMM models such as
CFMM AMMs. Impermanent Loss occurs when the original ratios of the assets de-
posited in a liquidity pool change, and the total value of the assets held by the user in
the pool are less than compared to what they would be, had they been held outside of
the pool. This leads to incurring what is called impermanent loss, a perceived loss due to
change in ratios of assets held, due to the nature of CFMM
Q such as UniSwap’s single pair
liquidity x · y = const or Bancor’s multi-asset liquidity i∈I xi = const. For a detailed
explanation on CFMM, see subsection 2.5.
For example, let’s take a pair of Cardano ADA and another hypothetical Cardano-native
asset X as assumption. We assume that at the time of providing liquidity to a CFMM
AMM style liquidity pool the price of ADA was $2.00 and the price of token X was
1, 000.00$. The initially provided liquidity, was therefore in form of 500 ADA and 1 token
X at the ratio of 1 : 1.
Now, having defined our starting state when contributing liquidity to the CFMM AMM
liquidity pool, we can define the impermanent loss as the result of asset ratios in the
pool changing. As the assets ratios change, we incur a loss in comparison to just holding
assets. The most optimistic scenario in the case of CFMM is the ratio remaining the
same, and hence putting the impermanent loss at 0%. However, as soon as the ratios
shift, we incur the unrealized loss, and in the extreme case of asset X increasing in price

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 41 of 88

25-times, while the price of ADA remained constant, we incur the loss of −61.5% (or the
loss of 16, 000$ given the initial investment of 2, 000.00$. All those losses are defined in
percentage-terms in Figure 17 and in absolute USD terms in Figure 18.

Figure 17: Percentage Divergence of Impermanent Loss on Initial Investment of $1, 000.00
at valuation of $2.00 and 1 Token X at valuation of $1, 000.00.

Figure 18: Divergence of Impermanent Loss in USD on Initial Investment of $1, 000.00
at valuation of $2.00 and 1 Token X at valuation of $1, 000.00.

It is worth noting that, as the name indicates, the loss is impermanent, meaning it is not
yet realized. If the liquidity pool would return to the original asset ratio there would be no
impermanent loss. Impermanent loss is realized in the moment of withdrawing liquidity
from the pool; once withdrawn, it is realized and can never be changed. However, due to
the highly volatile nature of crypto assets, keeping liquidity in the pool longer leads to
an increase in impermanent loss in the majority of cases.
Hence, we observe that changing initial CFMM AMM asset ratios causes impermanent
loss, but what causes the ratios to change? On the surface, the answer is simple, the price

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 42 of 88

divergence between assets being part of the pool leads to different ratios. Price itself can
be modelled using Geometric Brownian Motion (GBM) with drift and volatility; as such,
we can classify the below cases, in order from the worst to the best, in terms of loss due
to impermanent loss:

1. A cryptocurrency asset with a fixed max supply or with a burn mechanism (the
decrease or fixed supply leads to natural price appreciation over time), paired with a
fiat stable coin (which aims to achieve 2-4% over year level of inflation). Such a pair
is negatively correlated by design and will result in consistent loss of liquidity pair
value. For a liquidity pool open for an extended period of time (multiple years),
this can get very severe due to the fact that most CFMM move along a x1 line,
meaning that initially the loss is small, but then becomes very sharp. Given the
nature of this pair being negatively correlated, it is bound to go to 0, accelerating
along the way.
2. Cryptocurrency assets with misaligned drift5 . For instance, a new project with a low
starting price and high potential, and an established currency such as a blockchain
primary asset (e.g., ADA for Cardano), with high market cap and price stability.
The initial drift of the novel project and much lower market cap will lead to high
misalignment, and hence getting into the far parts of x1 impermanent loss causing
curve.
3. Cryptocurrency assets with similar drift. For instance, well established projects
such as Uniswap token and ETH. This kind of pair, due to relatively aligned drifts
and ecosystems, means a low impermanent loss, even smaller due to the fact that
1
x
gives very small impermanent loss within the local vicinity of the initial supply
ratio.
4. Assets reflecting the same underlying price, for instance stable coins of the same
currency, e.g., DAI, BUSD, USDC, and USDT. As those assets basically have ex-
actly the same drift, there is no impermanent loss present, only local divergence
to volatility σ between different pairs. That, combined with x1 impermanent loss
dynamics curve, means this is not noticeable, except in the case of black swan event
events.

6.2 Reducing Impact of Impermanent Loss

As we observed in subsection 6.1, impermanent loss comes from:

• shift in the initial ratio of tokens alongside 1

x
or other liquidity making curve (change
in the ratio of the provided assets);
• long time exposure allowing for significant shifts with high losses to occur;
• lack of compensation to market maker for taking on the risk of potential losses;
• lack of utilisation of external liquidity sources to rebalance internal value shift;
• inability to express the desired source and target liquidity pool asset ratios.
5
Drift is the trend along which the price moves, where volatility is the local divergence from the mean.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 43 of 88

Therefore, by addressing the above issues, one can eliminate completely the effect of
impermanent loss. This is achieved by using sophisticated modeling to asses the perceived
asset value and price dynamics, converging to desired assets ratios, the ability to start
from the preferred liquidity pool asset ratios, and dynamic market maker reward scaling
for providing liquidity in the market stress conditions, leading to a market net neutral
position.

6.3 Programmable Source and Target Asset Ratios

We start by tackling the issue of changing token ratios. In the majority of CFMM AMM
liquidity pools, all tokens are provided in the initial ratio of 50:50, which is the starting
point that achieves the highest impermanent loss. On the opposite side, 100:0 achieves
0% impermanent loss but provides very little liquidity. The ideal system would allow
the user to start with any initial liquidity and specify the desired target liquidity ratio
to allow the market maker to reflect the desired outcome. For instance, start with 100%
token X and using ratio shifting model end up with 80:20 of ADA to token X. The further
the maintained ratio is from 50:50, the lower the impermanent loss, as shown in Figure 19.

Figure 19: Liquidity Provision Asset Ratios Impact on Impermanent Loss

Not only does this help to reduce impermanent loss and converge onto the market maker
desired token allocation ratio, it also enables a series of interesting mechanisms, for
instance:

• Let´s take a new project launching its token X that wants to both bound liquidity
and to collect development funds. It could start with a 100% X based pool and
converge on the desired liquidity ratio of 20:80 (20% token X and 80% ADA), once
the pool stabilises on the 20:80 ratio, the project can withdraw 60% of liquidity
in form of ADA to finance the project development. Then, we would have 20% of
the initial token X provided into the smart contract and the 20% remaining after
withdrawing 75% of all ADA, leaving bounded liquidity with a 50:50 ratio and the
value of 40% of the initial supply.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 44 of 88

• Balancing the risk appetite in holding the assets by using, for instance, Modern
Portfolio Theory (MPT)[27, 26] or other portfolio allocation techniques, taking
risk / volatility as a parameter to construct liquidity pool with the desired risk to
expected revenue profile;

• By allowing every user to provide liquidity in the form of a different ratio, market
participants are able to reflect their true belief about the assets provided into the
liquidity pool.

6.4 Realistic Liquidity Supply Curve

As mentioned, the most popular CFMM AMM model is unrealistic and does only reflect
market conditions well due to arbitrage adjusting its value to the actual market conditions.
For instance, a CFMM with the formula x ∗ y = const composed of ADA and PIZZA,
where the current market price of PIZZA is 10 ADA per pie, assumes that the below 3
events have exactly the same probability density (likelihood of occurring):

• PIZZA costing 0.000001 ADA;

• PIZZA costing 10 ADA; and

• PIZZA costing 1, 000, 000, 000 ADA.;

As we can clearly see, the above probability distribution does not reflect reality, and
unnecessarily distributes TVL equally across the entire (0, ∞) domain.
In QuantFi, sophisticated models have been developed to precisely define price distri-
bution and dynamics and predict the market conditions, those methods are most often
based on a Geometric Brownian Motion (GBM) defining stochastic process St , following
GBM as

dSt = µSt dt + σSt dWt

where

• Wt is Wiener process (or Brownian motion);

• µ is drift;

• σ is volatility.

Using stochastic processes, we can predict and model realistic price probability distribu-
tion, and create liquidity models concentrating around moving drift µ, provided in a form
of oracle, and higher distribution moment:
q PN
(xi −µ)2
• volatility σ = i
N
– expected deviation from µ;
PN
(Xi −X)3
• skew µ
e3 = i
(N −1)·σ 3
- asymmetry around the mean µ;

• kurtosis κ = µ4
σ4
– tailedness;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 45 of 88

• higher moments in the case of high quality historical data available.

Figure 20: Stochastic Price Model.

Figure 21: Stochastic Process Reversal to the Mean.

Maladex protocol will enable a selection of liquidity pool models:

• stochastic pool model as outlined above;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 46 of 88

• legacy and custom formula models such as CFMM6 ;

• spot pools, pools converging to specific asset ratios according to the pricing formula.

The reason for all prices to be geometric in nature is the relationship between ratios and
logarithm, and ratios being how we express dynamic systems (in relation to themselves).
Unfortunately,
Q the majority of market making formulas have a constant form such as
x ∗ yconst, i∈I xi = const or similar, leading to high inefficiency in the pool price
discovery process. You can model any process iteratively with the worst model; given
a step is small enough, it will eventually converge, but in practice it gives rise to such
issues as:

• slippage – the execution moving the price significantly leading in overpaying for the
transaction;

• inertia – the pool is unable to update its price due to the astronomic size of the
total value locked, having single trades with very little effect on the price expressed
by the pool;

• unnecessary expenditure of energy – a lot of energy is wasted where it is not needed

(impossible scenarios), taking much longer to converge on the actual price. Yes,
arbitrage bots fulfill a role here of price balancer between exchanges, but there is a
significant lack of arbitrage between any assets other than BTC and ETH. People
who exchange cryptocurrency constantly are certainly aware of these often gross
pricing discrepancies for such liquid assets, as for instance Cardano on different
exchanges.

Maladex protocol will support models suited to different cryptocurrency asset types and
families of quantitative models to express the automated liquidity. The protocol will
support a rich family of quantitative models for minting liquidity pools, giving information
on what is the ideal pool in a given case and its parameterization, as well as allowing for
dynamic on-chain updates of the liquidity pool parameters via programmable swaps.

6.5 Algorithmic Liquidity Pool

The ability to model the price process well, as outlined in subsection 6.4, is one important
concept to reflect the price effectively, another is to react to changing market conditions.
Stochastic process will be able to update its state by the usage of price oracles, while
there are additional parameters we can update based on market dynamics:

• market maker fee – during the period of high traffic or rapidly changing price
distribution, the fee collected by market maker should increase; this is a fee for
6
Despite the fact that legacy liquidity formulas are much less effective, we predict that there might
be some uses, or users who will simply want to create them, and the premise of programmable swaps is
customizability, hence it only makes sense to include one of the most popular models as an option as
well.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 47 of 88

providing liquidity (service) in a market stress scenario and potential insurance

against black swan events7 ;

• yield farming – depending on the available liquidity in the market versus demand
for the liquidity from market takers, the rewards for providing liquidity will increase
in the form of MAL. As part of its role as unit of account MAL fees dynamically
adjusts to the market conditions, incentivizing action where it is needed.

6.6 Offsetting Internal Risk with External Liquidity

Evolving market conditions might cause the liquidity pool to incur impermanent loss. The
Maladex protocol reduces impermanent loss through the stochastic processes governing
it. The processes dynamically adjust fees and rewards, to result in a risk neutral liquidity
pool.
As pools are moved out of the true market price balance, an arbitrage bot could use this
opportunity of price difference between a specific liquidity fragment on Maladex and on
external DEXs, and will create transactions with it, resulting in additional revenue for
Maladex market maker and at the same time reverts the fragment price mean µ to the
perceived market price.

6.7 Minting Sound Liquidity Pools

It is very important for the market maker to pick assets to put into a liquidity pool that,
given the pool model, will minimise any potential losses while maximising the exposure
to money making opportunities. One wants a pair that is frequently exchanged, but at
the same time one that is able to maintain its inherent value over time, a property that
emerges from the specific assets in the pool and the model of liquidity that the pool uses.
An exchange such as the Maladex platform is perfectly positioned to aid users in creation
of sound liquidity pools.
In the majority of cases, simple CFMMs such as x ∗ y = const and the creation of
negatively correlated pairs deteriorates the total pool value. The worst offenders are fiat
stablecoins in pairs with cryptos. Stablecoins are by nature inflationary and crypto is in
its nature deflationary. This means that the liquidity pool from day one is a bucket with
a hole at the bottom through which your total rewards (rewards from providing liquidity
and liquidity locked in the pool) deteriorate, or in a good case, maintain current value
(which renders providing liquidity pointless).
Cardano now has a multitude of DEXs focusing on stablecoins and related liquidity pairs,
while the only valuable role of stablecoins is to bridge traditional finance with the world
of crypto. Once the bridge is crossed, there is very little reason to do anything with
the fiat. An exception might be perhaps in the rare conditions of very soundly designed
stablecoins such as Djed, which compensates swap of floatable for stable, but in the case
of Djed, it only makes sense to collateralise Djed with ADA, not to create liquidity pools
of Djed fiat and crypto.
7
This fee will have mean around µ = 0.3% similar to all other DEXs, but when the market will be
inactive it will decrease to incentivize trading activity, and when the market will be overheated and risk
will increase, it will be defined in such a way as to compensate the market maker for the risk taken.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 48 of 88

Maladex platform will provide information to the user derived from on-chain and external
data (via easy to understand user interfaces), performance metrics, aid in providing
quantitative feedback on the pool parameters selected by the user, and help to optimize
them to achieve the user’s goals. The visual cues will be especially helpful when trying
to understand the impact of parameters on the capital efficiency and the risks involved
(e.g., automatically alert if the pair reserves are negatively or weakly correlated). This
way market makers will be able to make far sounder decisions. Maladex Protocol further
incentives sound liquidity nodes with platform tokens, thus rewarding sound decision
making.

6.8 Fragmented Liquidity of AAMM

Each AAMM submission is a separate programmable swap, i.e., a separate EUTxO. This
fragmentation achieves the highest throughput and minimal memory requirement for
performing transactions. Security and soundness of the protocol is achieved via redeemers
and validators.
This is in stark contrast to all other liquidity providing models, where the design of
liquidity pools is very simplistic. It is easiest to design a pool with global memory, but in
the case of EUTxO account model it leads to 1 transaction per block (20s), or optimization
running into memory issues (16kB per transaction and 65kB per block). Doing it the
optimal way, utilizing Cardano’s EUTxO architecture, achieves high throughput, low
transaction fees, and opens a lot of new possibilities not available to other DEXs before.
The fragmentation of the liquidity protocol leads to a series of very desirable, naturally
emerging properties. If one shifts from monolithic blocks of liquidity to fragmented
nodes reflecting separate market making decisions, that naturally leads to the emergence
of geometric pricing models.
Uniswap v3, by introduction of concentrated liquidity and indexes, observed the same
issues[7]. Users specify liquidity in the ranges they think are good market making ranges,
and provide liquidity only within this range. As a result from all concentrated liquidity,
a geometric pricing model emerges.
It is a natural consequence of aggregation of multiple samples, each allowing for the
individual expression of market sentiment. This is due to CTL (Central Limit Theorem8 ),
which states that given a large enough set of random samples, a normal distribution
naturally emerges.
When we fragment the liquidity into separate nodes, all kinds of geometric distributions
naturally emerge, giving very efficient and market reflective pricing and liquidity models.
When combined with the Maladex domain specific language , it allows for the expression
of market sentiments, and an emergence of true market distribution. Maker orders in
the liquidity pool express their belief of the market, and given a large enough sample,
naturally reflecting market beliefs.

8
Central Limit Theorem article on Wikipedia: https://en.wikipedia.org/wiki/Central_limit_
theorem

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 49 of 88

7 Yield Curve
Yield is the product of engaging funds in a revenue producing activity, the higher capital
efficiency and the lower risk the better. The Cardano protocol and Maladex platform
have additional riskless sources of yield. We outline those yields in this section.

7.1 Yield Farming

Yield farming is providing financial incentives for users to take specific actions. On the
most basic level, yield farming would aim to increase the overall market efficiency, for
instance increase for pools and pairs lacking liquidity, offer higher rewards from minting
scarce options and synthetics, and incentivize the creation of high-quality DeFi educa-
tional content.

7.2 ADA Staking Rewards from Smart Contracts

The architecture of Cardano’s Proof of Stake (PoS) protocol and EUTxOs means that
it is possible to delegate ADA locked into smart contracts to pools, in order to earn
additional yield.
Any capital that is not actively used or required for a specific action can be delegated and
earn staking rewards. Hence, it also provides a nice benchmark for all investments. If
any investment is not able to generate revenue higher than the current Cardano network
ROI9 , then it is not worthy of engagement in, unless it provides hedge against risk.

9
Cardano staking reward act as network’s internal risk-free rate r.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 50 of 88

8 Emerging Properties of Programmable Swaps and

AAMM
The fragmented design of programmable swaps and mechanics built into AAMM lead to
the emergence of a series of very desirable properties:

• Fragmentation leads to geometric price distribution due to the Central-Limit Theo-

rem applied to the sample of investors; geometric price distributions are important
because they represent the market accurately, and quickly arrive at new stable
states when the market conditions shift.

• Freedom from impermanent loss due to all mechanics implemented into AAMM

• High capital efficiency, meaning higher revenue generating potential on the same
unit of money.

• Increased market efficiency, giving investors access to the true asset prices and a
trove of investment educational content and data.

• Protocol adaptability to shifting market conditions.

• Means of risk control, helping to protect the investments made.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 51 of 88

9 Indexes
An Index is a financial instrument representing a basket of instruments, allowing users
to track the performance of a group of assets in a standardized way[8]. Index allows you
to invest into the entire market, market sector, or a group of similar assets, such as DeFi
projects, DEXs, NFTs, etc.
The main advantage of holding an index, as compared to single crypto assets, is that
they often guarantee much more stable returns, at the efficient market rate, without too
much risk exposure. Take for instance smart contract blockchains; instead of placing all
bets on one horse or spreading bets according to your own metrics, one might buy shares
of a smart contract top 20 index, gaining exposure to the entire smart contract market
with allocation to specific assets correlated to market capitalization. As such, investors
are betting on smart contracts being successful overall rather than any specific project.
In summary, the advantages of investing in an index, compared to manual portfolio
management are:

• wider market exposure, increasing the market coverage and reducing volatility by
the fact that basket is comoset of different assets,

• protection as in hedge against single asset failures, e.g., if one DEX fails to perform,
then others in the portfolio will absorb its share, hence in that sense the risk is
neutral as the index represents the entire sector.

Some of the most popular examples of indexs from the world of TradFi and DeFi include:

• S&P 500 - tracking composed 500 largest companies stocks, weighted by market
capitalisation and representative of the entire US economy;

• FTSE 100 - 100 largest companies listed on the UK exchanges, weighted according
to their market cap;

• (Ethereum) DeFi Pulse Index: https://www.tokensets.com/portfolio/dpi ,

tracking the performance of the 25 largest Ethereum DeFi projects according to
the composition formula (each project capped at 25

A good example of a similar project from the Ethereum blockchain is TokenSet, which
provides indexes for a range of Ethereum blockchain tokens, e.g., DeFi Pulse Index. In
this case, a trader buys a stake of the index that represents the market composition, by
the capitalization for the specific group of assets (e.g., top 100 cryptocurrencies, top 25
DeFis, top 10 smart chain cryptos, etc.). The index is rebalanced at a constant interval
(typically 3rd week of every month) and maintains specific index properties (e.g., in the
case of TokenSet they limit max position to 25%).

9.1 Index Balancing

Asset prices and market capitalization at the time of rebalancing would be ingested from
on-chain Oracles, either based on:

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 52 of 88

• a centralized on-chain Oracle provider taking data from price trackers such as Coin
Market Cap, CoinGecko, and such data providers as CoinAPI, or

• from purely on-chain data build using Maladex protocol pricing models that should
reflect the true asset price on-chain in a trustless manner.

The index rebalancing can be achieved with one of many existing centralized price Oracles
and the typical index rebalancing of every 3rd week of the month.

9.2 Index Categories

Maladex is a Platfrom that allows for the deployment of the follwing.

• financial (fungible token) indexes - investing in fungible tokens representing different

projects shares / internal currencies;

• NFT indexes - index composed of NFTs bought and sold based on the current price,
valuation, and market sentiments, starting with art, but with the growth of NFTs
in the future potentially also representing all the deeds stored on the blockchain;

• Cardano delegation index - an aggregated way to delegate ADA via smart contract
to a portfolio of pools according to a specified condition, e.g., single stake pool
operators, ecosystem developers, mission-driven pools, etc. Each pool considered
for the index would need to meet the block production and reward requirements.

9.3 Cryptocurrency Index

The index protocol would consume on-chain price Oracle, reserve, and other data, accord-
ing to the index specification, and each cycle (T = 6 epochs = 30 days) would rebalance
the index according to the new parameters.
Moreover, CoinGecko tracks many different asset classes which is a great inspiration
for potential index creation. As such, we propose the development of the below crypto
indexes:

• Top 50 cryptos;

• Top 25 DeFi;

• Top 20 smart contract blockchains;

• Top 25 oracles;

• Top 25 specialized application blockchains (storage chains such as FileCoin or AR-

Weave or coverage networks such as HNT);

• and others.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 53 of 88

10 Synthetics
Mirrored (a.k.a. synthetic) assets are financial instruments deriving their value from the
underlying[15, 9, 12]. The underlying can be anything from real objects such as land,
art, or gold ownership to abstract such as commodities, indexes, stocks, equities, and any
existing financial instrument.
The name of this asset class is derived from their functionality, i.e. mirrored implies that
the asset mirrors the price behavior of its underlying asset which can be simple 1-1, but
also can represent an inverted (short), where the price is followed in inverse (e.g., a loss of
$10k in the underlying value is gain of $10 k in the inverted synthetic token). Synthetic,
on the other hand refers to a specific way of issuance of the mirrored tokens, namely by
not backing them with the actual underlying assets, but rather providing collateral acting
as insurance, providing the guarantee of the price. That collateral can be auctioned at
a discount to acquire back synthetic tokens and balance the value at risk requirement or
liquidated at the margin being returned as collateral to the token owners.

10.1 Advantages of Using Synthetics

It’s important to note a few distinguishing characteristics of synthetic instruments that
(often) make them more desirable to trade than the actual underlying instruments:

• Fractional ownership: minted synthetic tokens can be fractionally traded giving

access to them to low net worth individuals. Let’s take as an example a stock of
Berkshire Hathaway, worth at the 2021/04/27 market close $411,400.00. Creating
frictionless market access, gives much more fair access for financial betterment to
everyone regardless of their total net worth.
• Derivative financial instruments: synthetic tokens can use any formula to derive the
price from the underlying asset, from a simple reversed token, where for the case
of example, $10k loss in BTC price represents $10k gain in the reversed synthetic
BTC token. This effectively provides an easy mechanism to trade short (capital-
ize on the knowledge of the asset being overpriced) and leads to the creation of
much more efficient markets. Furthermore, derived instruments don’t stop with
reversed, but can be as well indexes, future contracts, options (deriving price from
the Black-Scholes formula), and much more. Covered calls and puts are the easiest
to implement using collateral.
• Liquidity: some assets pose a challenge to trade, e.g., physical assets or illiquid ones.
Synthetic token issuance, as it only reflects the price of physical commodities, makes
them easy to trade and in the case of illiquid assets, injects additional liquidity into
the system via the market making incentives proportional to the risk derived from
the exoticity of the underlying asset.
• No geographical boundaries: not everyone has equal access to financial markets;
from the brokerage requirements to restricted trading lists and actual market fric-
tions. A permissionless blockchain provides a platform for deployment of assets
available to everyone in the world and, in combination with synthetic instruments,
it gives access to trading them to a much wider audience. In turn, this creates a
more even field for all traders and injects additional liquidity into the market.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 54 of 88

• Low transaction costs: brokerages often charge significant fees for order execution.
In the case of exchanges or market makers, the provided options to trade in frac-
tional shares always come at a premium. The blockchain implementation both
provides transparency of what liquidity fees are, creates a peer to peer transaction
network, removing the intermediary and hence the cost, and provides much quicker
settlement mechanics.

10.2 Synthetic Token Interface

Synthetic’s token interface is composed of the following endpoints:

• Mint: provide collateral and the price oracle minting synthetic tokens.

• Burn: return synthetic tokens, burn them, and receive the stake of collateral rep-
resented by them.

• Trade: trade any token on the blockchain.

The critical functionality of synthetic tokens is the ingestion of asset price and the eval-
uation of whether the provided collateral is sufficient. If it’s not, then either the syn-
thetic tokens need to be repurchased at a premium using the collateral (which is simi-
lar to Maker’s Collateralized Debt Position - CDP) or the contract must be liquidated
[13, 14, 15, 16, 17, 9, 10, 11, 12].
It’s worth noting that the risk can be shared among a pool of similar or even all synthetic
instruments. The settlements would still reflect the price movements, but the risk can be
collateralized for the system altogether. [17]:

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 55 of 88

11 Financial Derivatives

11.1 Options
An options contract is a right, but not an obligation, to buy (or sell) the underlying
asset at a specific price at a specific time[25]. Options are great instruments to increase
the money making potential (as it allows the trader to take a larger exposure to the
price movements in the market), allow for risk control by providing inversely correlated
instruments (calls/puts), allow for structuring trades to bet on specific outcomes (e.g.,
take limited exposure to price increase via bull spread), and provide efficiency to the
market[25, 31, 32].
Options are priced using the Black-Scholes formula[25, 33]. Options derive their price
from the assumption that market dynamics follow a Generalized Brownian Motion and
the underlying asset volatility, free interest rate, and the option parameters themselves
(type - call or put, strike price, the expiration date).
Call option price C is defined as

C = St · N (d1 ) − K · e−r(T −t) · N (d2 )

where

• C – call option price;

• N – CFD of the normal distribution;

• St – current underlying price;

• K – strike price;

• r – risk-free interest rate (in the case of Cardano, this should be equal to the staking
reward);

• T – expiration date, where T − t is the time to expiry;

• σ − −volatility of the underlying asset;

ln(St /K)+(r+0.5·σv2 )·(T −t)
• d1 = √
σs · T −t
;
√
• d2 = d1 − σ s · T − t.

Put option price P is defined as

P = K · e−r(T −t) · N (−d2 ) − St · N (−d1 )

where all symbols including d1and d2 are the same.

We propose 3 option contract implementations:

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 56 of 88

• Covered options, where the option writer owns an equivalent amount of the under-
lying security. This means that the underlying is locked into the contract when the
covered call option is minted. This provides the guarantee that the option can be
exercised as per specification (according to the option type - European, American,
Bermuda and the price oracle). For covered option creators, it generates income in
the form of options premium.

• Collateralized option (synthetic instrument approach) where the collateral is pro-

vided to hedge the underlying volatility to around 200% the underlying value and
be subject to collateral rebalancing.

11.2 Advantages of option trading on the blockchain

Usually, options are written by big institutions (e.g., banks or large market makers) and
sold in bulks of options for 100 underlying shares each. Not only does this limit the
liquidity and who issues the option contracts on the market, but it also limits access to
buying option contracts (as they need to be bought in bulk of 100s and they carry the
corresponding risk with them). Not to mention, the cryptocurrency market is much more
volatile than the stock market, and taking the same exposure doesn’t make sense for a
lot of crypto assets.
Trading options on the blockchain provides 2 categories of benefits - to the market, by
increasing its efficiency and the number of instruments that can be traded, structuring
risk, hedging, and to investors, as doing so on the blockchain has a series of benefits:

• No middle man - options are minted by people on the blockchain and available to
everyone else on the blockchain. Thus, the execution fees are low, and the premium
is only paid for the service provided by the writer and the risk that they take;

• Fractional shares - options can be traded in much smaller units and even in fractions;

• Cheaper settlement - settlement is automatic via smart contracts;

• Full transparency - as all the information is publicly available on the blockchain.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 57 of 88

11.3 Option Trading Strategies

We further propose composite smart contracts allowing the purchase of option trading
strategies. A user would select the risk and profit profile, input strategy parameters, and
the smart contract would automatically structure using the required composites for the
strategy.
This adds a new innovative way of trading and makes option trading (usually very risky to
novices) available in a risk controlled manner (outlying the risk at value, the best/worst
potential outcomes, and for savvy traders, it just allows for an easy locking option strat-
egy).
We outline below a dozen of the option trading strategies that we’d expect to be intro-
duced on the Maladex platform as a follow up to the ability to trade options. This list
should evolve in the future and additionally welcome new strategy propositions from the
community, as long as they’re sound and the risk is easy to understand and display.

11.3.1 Bull Call Spread

Structure:

• Buy N calls, at strike price K1 with the expiration date T ;

• Sell N (the same number) calls, at strike price K2 where K2 > K1 , with expiration
date T (the same expiration date).

This type of vertical spread is used when the investor is moderately bullish on the under-
lying asset and protects themselves from the maximal potential loss, but also caps the
total profit if the contract expires above the strike price K2.

Figure 22: Bull call spread with strike prices K1 (low strike price) and K2 (high strike
price).

11.3.2 Bear Put Spread

Structure:

• Buy N puts, at strike price K1 with the expiration date T ;

• Sell N (the same number) puts, at strike price K2 where K1 > K2 , with expiration
date T (the same expiration date).

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 58 of 88

Similar to bull call spread, bear put spread is a tool for moderately bearish investors,
who wants to protect against the upside risk, and accept the potential limitation of the
income if the put expires deeper into the money.

Figure 23: Bear put spread with strike prices K1 (low strike price) and K2 (high strike
price).

11.3.3 Long Straddle

Structure:

• Buy N calls, at strike price K with the expiration date T ;

• Buy N puts, at strike price K with the expiration date T .

Long straddle does not provide risk protection, however it´s a bet on volatility, i.e., that
the underlying will move significantly off the strike price K, but without knowing in which
direction.

Figure 24: Long straddle with strike price K.

11.3.4 Long Strangle

Structure:

• Buy N out-of-the-money calls, at strike price K with the expiration date T ;

• Buy N out-of-the-money puts, at strike price K with the expiration date T .

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 59 of 88

Figure 25: Long strangle with strike price K.

11.3.5 Other Option Strategies

There are many other option trading strategies that can be structured from simple call
and put options, once those are available on the platform. We will provide the description
on the below strategies and more in the near future.
To mention a few other option strategies:

• Short Straddle,

• Long Call Butterfly Spread,

• Barrier options,

• Long term option,

• Calendar (Time) Spread,

• Collars,

• Iron Condor,

• Iron Butterfly,

• Fig Leaf,

• Long Call spread,

• Long Put Spread,

• Short Call Spread,

• Short Put Spread,

• Double Diagonal,

• .. and many more.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 60 of 88

12 Arbitrage
Arbitrage is the act of exploiting the market inefficiency between trading venues (ex-
changes). In the most simple case of the arbitrage, let’s say ADA is priced at $3 at
exchange A and at $2.95 at exchange B. The arbitrage bot will find this information and
will attempt to buy ADA at exchange B for $2.95 and sell it immediately at exchange
A at $0.05 profit per ADA. The act of arbitrage moves the price, as after the execution,
the price at the exchange B will be higher (e.g., $2.98) due to purchasing the discounted
ADA, and on exchange A it will be cheaper due to selling it (e.g., $2.98).
As long as there is a price difference between exchanges, so that when these trades
are executed they are more profitable than the execution cost, arbitrage is possible.
Arbitrage on exchanges will lead to them converging on the central limit value (true
market sentiment from the aggregated exchanges) of the asset price. The arbitrageur
earns profits by buying at a discount from exchange B and selling at the premium at
exchange A, hence both exchange B sellers and exchange A are the source of income for
the arbitrage bot.
Finally, the mentioned example is the most basic case of arbitrage (pure arbitrage), there
are a multitude of arbitrage styles, including statistical, where arbitrage is based on the
market movement predictions and the average expectation of reward10 . Arbitrage leads to
efficient price discovery, but at the cost of market participants (both takers and makers).
Information about the makers and takers of asset prices is spread across multiple ex-
changes; this leads to such gross discrepancies as the 3rd largest crypto (by market cap)
sometimes having 10-20 cents difference between exchanges. This gets only more extreme
for smaller cap coins, as while Bitcoin, Ethereum, and to a lesser extent Cardano, are
actively arbitraged, the smaller caps are deemed not worthy of running arbitrage bots.
This creates market segregation between exchanges and a lack of information flow. Arbi-
trage is a valuable and necessary information exchange mechanism and leads to efficient
price discovery and convergence to the same price +/- delta delays on participating ex-
changes.
The Maladex protocol with its fragmented liquidity, where each fragment can act as
supplied funds to a swap order on any Cardano-native DEX, will be able to capitalize
on inefficiencies on all other exchanges. Employment of arbitrage bots in liquidity pools
for ratio rebalancing and via arbitrage vaults means users can generate higher profits
compared to other platforms.
In the EUTxO model and script validation mechanism model, such arbitrage can be in-
corporated into a liquidity protocol. It will put market makers in the role of performing
additional transactions (and earning additional fees), for takers it will provide market
efficient price reflecting the true market price at any given time, and for the other appli-
cations it will provide invaluable on-chain price oracle data, not approximating the true
price, but having on-chain arbitrage proof of it. Such on-chain price Oracle does not have
any lag, meaning it is much more efficient than any other on-chain price Oracles (it is
the optimal state), only subject to lack of liquidity not allowing for convergence of price
in the market.

10
You can explore a few additional types of arbitrage by checking this Wikipedia article: https:
//en.wikipedia.org/wiki/Arbitrage#Types

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 61 of 88

13 Oracles
Maladex protocol is going to make use of oracles including pricing information and market
data. A series price and trading indicators, high quality moments of distribution (based on
stochastic modelling), and additional financial indicators such as Greeks (risk sensitivity
parameters)11 .

11
Greeks are a fundamental signal for trading options will eventually be available to users. You can see
the list of all (financial) Greeks in Wikipedia article, here: https://en.wikipedia.org/wiki/Greeks_
(finance)

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 62 of 88

14 Risk Control
Risk is an inseparable part of any investment, the most optimal trading strategy is always
the one that has higher revenue generating potential, without introducing additional risk.
In TradFi there is a plethora of measures to classify the level of risk such as VaR (Value
at Risk), volatility or potential loses. In addition, there are mathematical tools to classify
how good a given trading strategy is, taking the risk into account, such as Sharpe ratio,
and there are portfolio management theories that help to optimally allocate the capital,
given an acceptable level of risks, such as MTP (Modern Portfolio Theory).
Maladex Protocol design allows for the incorporation of risk control tool. Before we get
into the solutions, we shed light on why people take risk in the first place and what the
price of risk is.

14.1 Why do people take risk?

Risk implies the uncertainty of the outcome, where the outcome itself has some perceived
value. In light of the volatile nature of the market, high-risk investments are usually
those with the potential for high returns (although high risk does not imply high return
potentials). One would expect that in an efficient market (see EMH), the price would
reflect the risk involved. Unfortunately, this is often not true, owing to the inefficiencies
in the market. One usually makes returns by taking the risk; hence people take risks to
make profits.
Each investor may have a different risk appetite, and as such, one would construct a
portfolio with an expected risk level σ acceptable to the investor. This can involve a
mixture of high-risk assets, secure investments and also optimize for additional factors,
e.g., asset relationships to each other in the basket (see hedge).
The fact that risk is inseparable from investing does not imply that one should do nothing
about it. When actively monitored, one should adjust their position in the market due
to changing conditions and reevaluate their investment.

14.2 Market Maker Risk Compensation

The Maladex platform will aim to use information related to risk such as volatility to
dynamically change platform fees based on risk to market makers providing liquidity in
the risky assets.

14.3 Dynamic Risk Compensation

One can derive the current risk from the asset volatility (price fluctuations). Therefore,
the stronger and more rapid the fluctuations, the higher the risk. We propose to dynam-
ically adjust trading fees that go to market makers, based on volatility, to compensate
for the current and implied risk level.
This is an efficient risk prevention mechanism, as it helps to cool the market, due to
execution fees (going directly to market maker), compensating for on-going risk. Also, to
some extent, during normal market conditions (e.g., when the volatility is not related to

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 63 of 88

a big announcement where the fundamental value of the asset changed significantly, e.g.,
dropped to 0), this mechanism is able to put market makers in a risk neutral position
(remove risk from providing liquidity).
Finally, in the extreme case, of the disclosure of new information in the market (e.g., a
project turning out to be a scam and token fundamental values changing to 0 in a blink
of an eye), risk management would be covered by market maker fuses, outlined in ??.
Be that as it may, there is one more thing that can be done to protect market makers
in the face of an extreme event, where one of the assets suddenly loses all of its value,
namely arbitrage.

14.4 Token Trust Scores & Whistleblowing

All Cardano-native assets have a unique PolicyId which allows the unique identification
of the assets. PolicyId is often used, on NFT trading platforms, to confirm that a specific
image belongs to the official collection. That same mechanism can essentially be used for
approved tokens, confirming that the token in question is an official one.
However, there is much more information that can be attached to PolicyId such as

• mint information (how many tokens have been minted, is the policy locked, etc.);

• information about the project (website, white paper, etc.);

• information provided by the community.

The last point, information provided by the community, can be especially effective at
performing the role of project information aggregation, as users share:

• good signs about the project;

• suspicious activity;

• and even in extreme cases, whistleblow an on-going scam.

The information provided by the users could be compiled down into a trust score and
provided alongside the tokens, and all objective information could be included alongside
the score for each user’s review.

14.5 Quantification of Risk

There are many ways to quantitatively assess the risk[131]. The most popular methods
are:

• Markowitz’s
P Modern Portfolio Theory (MPT) with the
PNportfolio’s returns defined
as µp = N 2
PN
w µ
i=1 i i and portfolio’s variance as σp = i=1 j=1 ij i wj. Portfolio
σ w
might include a single liquidity pool with the desired risk profile of σp .

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 64 of 88

• Capital asset pricing model with µi = Rf + βi (µm − Rf ) and βi = σσim

m
where Rf is
the risk-free rate of return (in the case of Cardano the current ROI on staking in
a typical pool), µm is expected market return, βi is the beta coefficient of asset i,
and σim is covariance of asset i in the market m and σn is the standard deviation
of such market.

• Value at Risk (VaR) which simplifies the question of potential loss compared to both
above models to the question “How much can one expect to lose, given cumulative
probability ζ, for the given time horizon T , defined as F (Z(T ) ≤ VaR) = ζ, where
F is cumulative distribution function, Z(T ) = S(0) − S(t) is the loss for an asset
S at time t, and ζ is a cumulative probability function associated with threshold
value VaR, on the loss distribution of Z(t).

• Coherent risk measures such as CVaR (Coherent VaR), defined as CVaR = E[Z(T ) |
Z(T ) > VaR] and copulas.

All the above risk metrics can be used both in portfolio construction and rebalancing.
It is worth noting that, for the sake of risk management, portfolio can refer to almost
anything, most importantly including liquidity pool where it is used to actively balance
the risk of AMM providing liquidity to the market. Note that this mechanism offers
means for automated risk management (for the risk appetite defined by the user) using
programmable swaps and Algorithmic AMM.

14.6 Moment Indicators

As outlined in section 6, moments of price distributions are the base for all the derived risk
metrics and multiple trading indicators. As such, it is important to visualize the realized
and implied volatility of the assets in the market, and both give the user visual indication
of historical volatility, as well as allow them to define their own balancing mechanism in
programmable swaps where desired. Moment based indicators provide the most verbose
building components for the creation of automated risk management strategies.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 65 of 88

15 Maladex Protocol Settlement Layer

Maladex aims to become the execution engine (settlement layer) for the financial world.
programmable swaps will perform automated trading. We will keep all settlement on-
chain, the protocol needs to solve a unique set of challenges in decentralization, memory
throughput, and execution costs.
The largest DEXs, such as Uniswap, are able to reach a volume of 100,000 swaps per
day, which as blockchains go is an impressive number, but this happens at a very high
average execution cost, currently around 30 - 100 USD per transaction, and fades in
comparison to TradFi, where millions of trades are executed every minute. We present
the comparison of exchange trading volumes in Table 1.

Table 1: Exchange Amount of Transactions per Day, Week, and Month.

Average Transactions Per
Exchange Exchange Type
Day Week Month
Uniswap v2 Ethereum 50,288 352,000 1,508,640?
DEX [138, 139]
Uniswap v3 Ethereum 17,858 125,000 535,740
DEX [138, 139]
Uniswap Total Ethereum 68,149 477,000 2,044,470
DEX [138, 139]
SushiSwap Ethereum 17,869 125,080 536,070
DEX [138, 139]
Binance Crypto CEX 473,000 [141] 3,311,000 14,190,000
Nasdaq TradFi CEX 26,390,296 131,951,480 580,586,512
[142]

15.1 Maladex as Scaling Layer for Cardano-native Projects

The extensible nature of programmable swaps and our concurrency solution reaching
theoretical limits of scalability, means that in the future all Cardano-native projects will
be able to utilize Maladex protocol to scale their own protocols. For instance, let’s take
minting NFTs, which currently requires all creators to prepare a scalable mint process
and additionally might exceed the current mempool size (equal to 2 blocks, i.e. 2 · 65kB).
This could be addressed by creating a mint contract using a programmable swap (swap
X ADA for randomly selected NFT using random number generator (RNG) oracle). Such
a contract would be easy to implement and could even be done using a graphical user
interface (GUI). We would utilize a hydra head scaling solution and optimal concurrency,
leading to the creator being able to focus on the art and community building, while
outsourcing the minting process to Maladex protocol, where they would benefit from
available scalability and low transaction costs.
This is just a single example, but we see this model extending to many more use cases.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 66 of 88

16 High-Frequency Data Lake & Lab

Strategy development, testing, and execution requires quality sources of data.

• on-chain price oracles based on asset valuation model as outlined in section 6; those
oracles will provide high quality and accurate indicators from asset price to a wide
range of indicators, most importantly moment-based ones, fundamental for writing
algorithmic trading strategies;

• data lab – data warehouse and live data streams interactions, generalized strategy
writing API, and is a convenient way to perform backtesting and simulations.

Access to data will involve payment in MAL utility token for the higher levels of con-
sumption.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 67 of 88

17 On-chain Hedge Fund

Maladex’s programmable swaps provide a unique mechanism for implementation of an
on-chain hedge fund. Hedge fund refers to a pooled investment, taking advantage of so-
phisticated trading strategies and risk management techniques, in an attempt to improve
performance above the index benchmark.

17.1 On-Chain Portfolio Managers

Maladex will create a market for trading strategies by enabling the publication of one’s
programmable swaps and setting a performance based fee, for instance 1% of all revenue
generated using the strategy. PMs are going to be incentivized to create and publish
trading strategies, because of the performance based fee they are going to be able to earn,
and platform users will be incentivized to lock funds into strategy vaults by being able to
see strategies’ historical performance and PM’s rankings[24, 27].
PMs via Maladex platform will have access to sophisticated tools such as high-frequency
and quality data feeds, a backtesting platform, and a wide array of financial instruments
to implement strategies, from indexes and synthetics, to financial derivatives and options
strategies.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 68 of 88

18 DeFi Education Portal

We estimate that only about 3% of cryptocurrency holders actively engage in using DeFi
products. Moreover, approximately 80% of those users report that they do not under-
stand / would like to better understand impermanent loss, capital efficiency, and how to
quantitatively assess their decisions. What is more, the cryptocurrency market has one
of the weakest forms of EMH.
We see it as essential, next to providing sound investment products, to provide observa-
tions and orientation (see OODA – Observe-Orient-Decide-Act loop [130] to learn more
about this framework of thought) to the users of DeFi platforms, in order to be able to
take better actions and to create a feedback loop by visualising the available telemetry.
One of those 3 key components is the development of an incentivized DeFi educational
platform.
Maladex is going to create a platform for users to create DeFi-related educational content.
The views, engagement level, and kudos will then be used to allocate a portion of MAL
tokens to articles based on their popularity and usefulness. This is going to create a
financial incentive for people with experience in the field to create quality content, and
share it with the community.
On top of that, Maladex team is going to publish a series of free articles (not participating
in the outlined above token allocation) explaining many of the key concepts in DeFi.
In the future, we see the DeFi educational portal also being used for publication of market
research, potentially with different funding models other than for free (the creation of
high-quality research requires a lot of work, so a subscription-based model might be
introduced for experienced and high-volume publishers).
Finally, throughout the Maladex platform, users will be able to find references to this
educational content, to be able to get information when they need it the most. Some of
that research will be further published by the Maladex team for free, based on the unique
insights available to our DeFi platform, embedded in all kinds of market making, across
many financial instruments.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 69 of 88

19 Tokenomics
Maladex platform has a Cardano-native utility token called MAL. MAL token is shared
across all Maladex products and is the platform native unit of account which enables
access to platform functionality. MAL has a fully diluted supply of 42, 000, 000 and
approximately a 5 year gradual vesting period.

19.1 MAL Token Distribution

MAL token is allocated in the below percentages:

• 20% allocated to the team; with 4% vested every year, over a total period of 5
years, reserved for yearly team performance bonus;

• 10% reserved for the allocation before platform launch;

• 25% reserved for treasury liquidity bonding over the span approximately 5 years
(minimal amount of tokens will be bounded at launch only to guarantee the liquidity
of MAL tokens and the rest is going to be released over years);

• 45% incentivization rewards, 40% reserved for rewards for using the platform, such
as yield farming, DeFi educational content creation, etc., and 5% reserved directly
for the airdrops to incentivize and reward actions outside of the mentioned ones
(e.g., reward for moving liquidity from different exchange, reward for having used
the platform before date X, etc.).

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 70 of 88

Figure 26: MAL token allocation.

Team
Y1 Y2
Airdrops Y3
Y4
Y5
Rewards

Platform Rewards Private

Sale
Public

Liquidity bonding

Treasury

19.2 MAL Token Vesting Schedule

MAL token has the following vesting schedules:

• team allocation is going to be vested yearly; each year up to 4% of discretionary

bonus will be distributed. First distribution (year 1) will include an additional
lock-in period to ensure that the team allocation to minimise the effect on the
market;

• public allocation will unlock after on the platform launch;

• private allocation will have a vesting period;

• liquidity bonding (treasury) will unlock gradually to ensure sufficient liquidity of

MAL token in the market;

• platform rewards will unlock gradually and will be dynamic to incentivize specific
behaviors (providing new liquidity pairs, creating healthy liquidity, content creation,
etc.); we expect that it will take multiple years (about 5) for all rewards to be
distributed;

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 71 of 88

• airdrops will be applied in response to specific events, hence it will be less gradual,
but only small portions will be used at times.

MAL initial supply at launch will be around 4, 200, 000 (10%) coming from both private
and public allocation and potentially small amount (about 1%) of initial liquidity bonding.

19.3 MAL Token Utility

MAL is a Cardano-native fixed-supply utility token. MAL token derives its value in the
market from the perceived functionality by the users. This is a model with strong funda-
mentals as the utility guarantees healthy price dynamics as opposed to purely governance-
based tokens.
MAL token is currently designed to provide the below utility (but in the future might
include additional utility):

• native settlement unit of account for hydra head (layer 2) execution and pro-
grammable swap resources;

• computational credits for back-testing and access to high volumes of historical data;

• consumption of on-chain oracles;

• tips to DeFi educational content creators;

• locking a specific amount of tokens (without spending them) will give access to a
series of professional data;

19.4 Maladex Treasury

MAL collected from protocol utilization outlined in subsection 19.3 will go into Maladex
treasury, where they will be used to:

• fund protocol development and DeFi research;

• fund the platform infrastructure;

• be distributed to the platform users in the form of incentives (once the initial 5
year supply of reward allocation is exhausted the incentives will be funded from the
treasury);

• provide MAL incentives for creation of mathematically sound liquidity pools and
instruments;

• be used to fund the rewards distribution for DeFi educational content writers,

We project that healthy inflow of MAL tokens from the protocol usage and its reinvest-
ment into the platform will lead to a very healthy positive-feedback loop ensuring.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 72 of 88

Acronyms
AMM Automated Market Maker (AMM). 7, 12, 13, 15–17, 28, 40, 41, 43, 44, 64, 77,
78

CEX Centralized Exchange (CEX). 11

CFMM Constant-Function Market Maker (CFMM). 7, 10–12, 14–16, 40–44, 46, 47, 78,
80

dApp Decentralized Application (dApp). 24, 78

DCA Dollar-Cost Averaging (DCA). 26, 35, 36, 38, 74, 75

DDoS Distributed Denial of Service attack (DDoS). 26, 74

DeFi Decentralized Finance (DeFi). 1, 9, 10, 18, 49, 51, 68, 69, 71, 77, 80, 81

DEX Decentralized Exchange (DEX). 12, 17, 27, 39, 47, 48, 51, 60, 65, 74, 76, 80

DSL Domain-Specific Language (DSL). 27, 35, 36

EMH Efficient Market Hypothesis (EMH). 10, 11, 40, 62, 68

EUTxO Extended Unspent Transaction Output Model (EUTxO). 15, 17, 24–33, 35,
48, 49, 60

FPGA Field-Programmable Gate Array (FPGA). 10

HFT High-Frequency Trading (HFT). 9

ISPO Initial Stake Pool Offering (ISPO). 77, 80

MEV Miner Extractable Value (MEV). 73

NFT Non-Fungible Token (NFT). 23, 25–27, 51, 63, 65

P&L Profit & Loss (P&L, PnL). 27, 79, 81

PAB Plutus Application Backend (PAB). 24, 79

PM Portfolio Manager (PM). 35, 67

QuantFi Quantitative Finance (QuantFi). 9, 40, 44

SPO Stake Pool Operation (SPO). 39, 77, 80

TradFi Traditional Finance (TradFi). 1, 9–11, 15, 18, 35, 40, 51, 62, 65, 77, 80

TVL Total Value Locked (TVL). 15, 44, 74, 80

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 73 of 88

Glossary
account-based model A ledger model used by Ethereum, and majority of smart con-
tract enabled blockchains, where the global state is shared, and all operations are
applied sequentially, one after the other based on tips. Due to possible impact on the
ordering of transactions via tips it is prone to front-running and Miner Extractable
Value (MEV).. 40

Amdahl’s law The theoretical speedup limit in the latency of the execution of a task
at fixed workload that can be expected of a system whose resources are improved,
it also is used to define the theoretical limit of the system scalability due to im-
provements in concurrency and parallelism of the system. Amdahl’s law is defined
1
as Slatency (s) = (1−p)+ p .. 25, 27, 74
s

arbitrage Exploiting the market inefficiency between trading venues (exchanges); in the
most simple case of the arbitrage, let’s say ADA is priced at $3 at exchange A
and at $2.95 at exchange B. The arbitrage bot will find this information and will
attempt to buy ADA at exchange B for $2.95 and sell it immediately at exchange
A at $0.05 profit per ADA. The act of arbitrage moves the price, as after the
execution, the price at exchange B will be higher (e.g., $2.98) due to purchasing
the discounted ADA, and on exchange A it will be cheaper due to selling it (e.g.,
$2.98). As long as there is a price difference between exchanges, so that when these
trades are executed they are more profitable than the execution cost, arbitrage
is possible. Arbitrage leads to all included exchanges converging on the central
limit value (true market sentiment from the aggregated exchanges) of the asset
price. The arbitrageur earns profits by buying at a discount from exchange B and
selling at a premium on exchange A. Hence both exchange B sellers and exchange
A are the source of income for the arbitrage bot and are ”victim” of the market
inefficiency (in reality, they are not victims but simply prone to be scalped this way).
Finally, the mentioned example is the most basic case of arbitrage (pure arbitrage).
There are a multitude of arbitrage styles, including statistical, where arbitrage is
based on the market movement predictions and the average expectation of reward
https://en.wikipedia.org/wiki/Arbitrage#Types. Arbitrage leads to efficient
price discovery, but at the cost of market participants (both takers and makers)..
20, 60, 74

Automated Market Maker (AMM) A liquidity definition as a supply formula en-

abling automatic (autonomous - without the presence of maker) trades, preserving
certain properties such as path independence (the execution price is not depend-
ing on the history of transactions), and for trading without any active interaction
from the maker party (makers provide assets into the liquidity pool which are then
managed based on pool model / formula).. 72

benchmark Benchmark is an investment performance measuring tool, used to assess the

allocation, risk, and return of the portfolio[29]. Benchmarks are usually constructed
using unmanaged indices and exchange-traded funds (ETFs). Benchmark is selected
in a way to represent asset class against which one wants to compare the outcomes..
67

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 74 of 88

black swan event The black swan theory or theory of black swan events is a metaphor
that describes an event that comes as a surprise, has a major effect, and is often
inappropriately rationalized after the fact with the benefit of hindsight.. 42, 47

capital efficiency Each task has many ways of being performed, one way will consume
more resources (energy, time, etc.) than the other, hence the way which consumes
less resources and achieves the same result is more efficient. The more efficient use
of capital the better generated returns, the better user experience, and the better
the actual state of the market is reflected.. 1, 15, 18, 40, 49, 50, 68

Centralized Exchange (CEX) Classical model of an exchange where agents engage

into buying and selling via the intermediary (the exchange), as opposed to DEX
where investors face each other via the protocol implementation.. 72

concurrency The act of progressing on the same task by multiple agents at the same
time, which implies communication between agents. Concurrency speed improve-
ment is limited to the bottlenecks in agent communication and defined by Amdahl’s
law.. 73

Constant-Function Market Maker (CFMM) A liquidity provision formula that has

assets on one side and a constant on the other. The most widely known CFMM
is Uniswap’s v1/v2 x ∗ y = const [5, 6]. The critique of this liquidity formulation
include unrealistic provision range (assumption that it is equally likely to provide
liquidity (make the market) when the asset that costs $20 (e.g., a pizza pie) at price
of $0.000001 and $1,000,000,000. This inefficiency leads to high impermanent loss
and requirement for high TVL to avoid high slippage, which leads in turn to market
inefficiency and creating a huge potential for arbitrage.. 72

Decentralized Application (dApp) An application that runs on a Decentralized com-

puting system, such as Cardano blockchain.. 72

Decentralized Exchange (DEX) Non-custodial exchange model, where trades are ex-
ecuted directly on the smart contract blockchain. DEX does not have intermediary,
responsible for providing the service for exchanging assets, but rather it is performed
in automated manner by participants providing assets for the code to access (gov-
erned by code rather than company).. 72

Decentralized Finance (DeFi) Blockchain-based form of finance, removing the need

for the centralized intermediaries required to provide the service in the traditional
finance, such as brokerages, exchanges, and banks. In DeFi the role of intermediary
is replaced by smart contracts.. 72

Distributed Denial of Service attack (DDoS) A form of attack on the Internet in-
frastructure, where the attacker aims to exhaust the available resources of the target
by the shere volume of the traffic created. DDoS usually results in the targeted ser-
vice being unavailable for the duration of the attack.. 72

Dollar-Cost Averaging (DCA) An investment trading system employed to minimise

the impact of local price fluctuations on the average buy-in-price. A person fol-
lowing DCA trading system would purchase the same amount of the target asset

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 75 of 88

(e.g., ADA), at equal interval (e.g., every day), for the same amount of the source
currency/asset (e.g., USD). There are more sophisticated forms of DCA available,
for instant, weighted DCA in which the amount bought at the equal interval (e.g.,
every day / week) is adjusted by the weight parameter w computed from the mar-
ket indicators (e.g., divergence indicator indicating local maxima and minima). In
the case, of divergence indicator, the weight w would be higher near local minima
and lower at the local maxima, resulting on average in better performance of the
trading system.. 72

Domain-Specific Language (DSL) A computer language specialised to a particular

application domain, e.g., the language for defining Maladex programmable swaps..
72

Efficient Market Hypothesis (EMH) A hypothesis stating that the prices in the
market reflect all of the information available, in practice it is rarely true. A
straightforward example is Dogecoin rising to the top 10 and firmly remaining there
based on a series of tweets with no substance behind them. Even a rebuttal from
the Dogecoin development team did nothing to stem the rise. The market ignored
this information and irrationally valuated Dogecoin at a higher price. This does not
reflect the fundamental information available in the market and is an extreme case
of market inefficiency. In traditional finance, research would be published showing
its shortcomings, people would evaluate the study’s validity, and would short such
an overvalued company. If an asset were to be severely undervalued, a leverage or
long option would be bought, significantly magnifying the potential returns from the
efficient use of information. Under the Efficient Market Hypothesis, assets would be
priced at what they are truly worth, and there would be very little difference in the
asset prices between exchanges. On the contrary, market inefficiency is such a com-
monly occurring phenomenon that a whole field of study is dedicated to measuring
it and adjusting trading strategies based on it. Inefficiencies take root from lack
of information flow between exchanges (e.g., via arbitrage) and irrational decision-
making (modeled well by behavioral investing). This is not a problem without a
solution, it is merely the result of the environment emerging from inefficiencies of
the exchanges and the irrationality of the actors in the market.. 72

Extended Unspent Transaction Output Model (EUTxO) Accounting model used

by Cardano where all assets are stored in EUTxOs (boxes) that can be spent either
using private key (e.g., wallet transactions) and (Plutus) script (private key and
script are locked on the box). The wallet balance is the sum of all EUTxOs that
the wallet can spend. In the EUTxO model each EUTxO has to be spent when
used, hence can be only used once per block (20s).. 72

Field-Programmable Gate Array (FPGA) Programmable hardware circuit composed

of programmable building blocks and a hierarchy of reconfigurable interconnects al-
lowing blocks to be wired together. FPGA is programmed in hardware description
language (HDL) and once encoded behaves as an integrated circuit designed to
perform a specific functionality. FPGA are used to write programs directly into
silicon, significantly improving processing speed and allowing for many hardware
optimisations. The only step further is ASIC (Application-Specific Integrated Cir-
cuit) which requires a fab (factory configuration to produce ICs), hence in all the

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 76 of 88

special and often changing cases, prohibitively expensive (e.g., in HFT where the
algos are constantly updated and improved).. 72
front-running Process of using the existing knowledge of submitted orders (e.g., by
scanning mempool) to tip the exchange / protocol, with the aim to insert your
order (based on existing information) ahead of other orders, in order to capitalize
on this knowledge. For instance, spot a whale swap, tip the protocol to insert
transactions before in the block, buy at the cheaper price, allow the whale order to
move the market (whale pays more for the order), tip for 2nd order to be placed
after the whale order, sell immediately for higher price.. 73, 77

hedge Hedging is the practice of taking a position in an inversely correlated asset/market

to offset the risk taken by assuming position. Hedging might take many forms, but
always leads to reduction in the overall position risk. For instance, when shorting
stock using options, assuming also long position in the underlying.. 51, 62
hedge fund A pooled investment fund, often trading in relatively liquid assets, and
able to make extensive use of sophisticated trading, portfolio construction[24, 27],
and risk management strategies in an attempt to improve performance, especially
improve the performance above index benchmark, by generating alpha without
taking on more risk (i.e., achieving higher Sharpe ratio).. 35, 67
High-Frequency Trading (HFT) Sophisticated algorithmic trading methodology char-
acterized by high speeds (microsecond level), exchange co-location, high trade vol-
umes, and high order-to-trade ratios. HFT most frequently utilizes the local market
volatility and heavily relies on momentum-based indicators and derives its rev-
enue from the trade volume and being right more often than wrong approach,
compounded over millions of trades executed. HFT does not consume significant
amounts of capital, nor accumulate positions, or hold portfolios overnight, and as
such the results have some of the highest Sharpe ratios (reward to risk ratio) in
the market. HFT is often seen as a method of providing liquidity to the market as
most HFT algos will trade both sides tightly around the spot price. 72

impermanent loss Loss of capital when compared to HODLing due to inherent flaws of
liquidity pool formula. The worst offenders are ironically the most popular family
of constant-product pools such as Uniswap’s v1/v2 x ∗ y = const[5, 6]. It occurs
when providing a pair of assets into a liquidity pool, as the rations of x and y
change in the pool and the more the prices diverge, the bigger the impermanent
loss. At some point, around the 20-50% threshold (the range is large due to inherent
typical crypto volatility), actually holding the 2 separate assets outside of the pool
would have been a better investment. It is made even worse by the incentivization
of yield farming to create the most negatively correlated pools there are possible
(completely deflationary crypto such as ADA and completely inflationary fiat such
as USD). Those two assets will naturally diverge and create the impermanent loss.
In a sense, there’s a sizeable hole in the bucket through which the total value of
your deposited funds will leak. This is made even worse by many DEXs being either
only concerned with increasing the amount of transactions (in order to obtain more
fees from transactions) or being completely oblivious to this obvious mathematical
fact (DEX are generally built without regard to quantitative modelling). . 1, 10,
15, 16, 18, 40–43, 50, 68, 74

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 77 of 88

index An index is a method to track (and replicate) the performance of a group of

assets in a standardized way. Index typically measure the performance of a basket
of assets intended to replicate a certain area of the market.. 51, 67, 76

Initial Stake Pool Offering (ISPO) A novel project cost development financing model
utilising Cardano SPO. A project wanting to receive funds via Initial Stake Pool
Offering (ISPO) will raise variable fee to 100%, collecting all rewards from the block
production. In return, delegators receive airdrops of the project utility tokens, to
be used when platform launches to access its features.. 72

limit order A limit order is an order to buy or sell stock at a specific price or better.
A limit order ensures execution at the specified price (or better), but does not
guarantee how quickly the order will be executed, or if at all. Limit orders are ideal
when the primary goal is to execute at specific price, once that price is available..
20, 26, 30, 34–36, 38

liquidity Ease of exchanging one asset for the other without affecting the asset price,
at the perceived current market value. Naturally the most liquid asset category
are fiat currencies, and the most liquid among them USD, which can be exchanged
for many assets around the world; it is used in international settlements, and the
payments performed by individual organisations, regardless how large, do not affect
the price of USD.. 9, 10, 12, 15, 16, 27, 32, 40, 41, 44, 47, 48, 64, 77

liquidity pool A method of providing liquidity to the market via crowdsourcing it from
among many participants into one shared pool, usually in the form of AMM.. 14,
16, 26, 29, 31, 34, 38, 40–45, 47, 48, 60, 64

market maker A party providing liquidity to the market (e.g., via liquidity pool), either
by depositing assets into liquidity pool, usually constant formula pricing assets, in
the case of DeFi, or via sophisticated market making tools and algos in the case
of TradFi. Market makers create efficient market to transact in, ensuring good
experience of market takers.. 9, 10, 14, 15, 17, 30, 32, 35, 42, 43, 46, 47, 62, 63, 77

market order A market order is an order to buy or sell a stock at the market’s current
best available price. A market order typically ensures an execution, but it does not
guarantee a specified price. Market orders are optimal when the primary goal is to
execute the trade immediately.. 26, 28, 30, 31, 35, 36, 38

market taker Any party exchanging assets on the market, and hence taking the liq-
uidity from the market, or in other words taking the service provided by market
makers. They expect the exchange prices to reflect the true prices of assets, low
slippage (price change during trade execution), and ability to exchange assets on
demand.. 9, 10, 30, 32, 35, 47, 77, 80

Miner Extractable Value (MEV) A measure of potential profit available to miners

(validators, sequencers, etc.), arising from the ability to arbitrarily include, ex-
clude, and re-order transactions. For instance, the existence of MEV enables front-
running.. 72

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 78 of 88

model A mathematical description / definition of system, e.g., liquidity model defining

how liquidity pool models market dynamics in the response to trades. However,
even a poor model can be used for the purpose of modelling, but it will suffer from
many inefficiencies (lagging behind the actual state, being too sensitive / having
too high inertia, etc.). All CFMM AMM are poor liquidity models.. 40

money ≡ energy When modelling financial systems you can think of money as a unit
of energy, same as in physics, we can think of kinetic, potential, and other energies,
energy is potential to perform a certain amount of work; we can think of money
in the same way. Money has potential to be applied in the market, from lending
to businesses developing novel ideas, increasing market efficiency by providing liq-
uidity to exchanges, creating information flow via arbitrage, financial analysis and
research, to utilising available information to maximise returns given a specified risk
appetite. Energy can be applied in many ways. If we choose a very inefficient way
to do something e.g., growing tomatoes in a completely dark underground bunker,
we will expend a lot of energy. In contrast, we can put them in a greenhouse or
even on a balcony and make much more efficient energy use. Same in the capi-
tal markets, efficient models lead to high money utility (energy well spent towards
productive work), and productive work is the work that is usually associated with
rewards (e.g., business borrowing money will pay it with interest). Therefore, an
efficiently allocated liquidity will work very well even with low total value locked
and respond quickly to the changing market dynamics, bringing higher rewards per
unit of capital.. 40

Non-Fungible Token (NFT) Is a unique and non-interchangeable unit of data (most

often JSON) stored on the blockchain. NFTs are most often used to store images,
and do so via 721 standard defining the JSON template. However, NFTs have
many uses, such as storing wallet handles, and even such as in the case of Maladex
the actual code of programmable swaps.. 72

off-chain Everything that is not on-chain, most often implies off-chain code of dApp
smart contracts.. 7, 15, 19, 24, 25, 27, 38, 79

off-chain code Cardano introduces the concept of Turing-complete off-chain code, that
computes the necessary state update for the user taken action, and provides it to
Cardano wallet for submission onto the ledger.

In contrast to blockchains without safe off-chain component, it provides the ability

to write off-chain code in any existing programming language (as long as bindings
to Plutus exist, e.g., via an SDK), and provide any functionality without outsourc-
ing the high execution cost on the users (e.g., via fee mechanism).. 15, 17, 24, 27,
78, 79

on-chain Action executed by the validator nodes or an asset stored on the ledger.. 1,
14, 17, 19, 24, 27, 46, 48, 60, 65, 79

on-chain validator A part of smart contract code stored on Cardano ledger and used
to validate submitted transactions. In contrast to other blockchains where all smart
contract actions are performed on-chain, in the case of Cardano, a transaction is

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 79 of 88

prepared by off-chain code and validated if it is spendable and meets all spending
conditions, by block producing nodes, on layer 1.. 15, 17, 24, 27

parallelism The act of splitting a task into smaller, independent, subtasks, all of which
can be executed independently of each other, implying that there is no communi-
cation required between agents executing the subtasks.. 22, 25, 73

Plutus Application Backend (PAB) An off-chain dev tool allowing for the interac-
tion with smart contracts. PAB allows for interaction with external clients, such
as wallet front-ends, and acts as the intermediary between Plutus application, the
node, the wallet back-end, and end users. The purpose of PAB is to provide a
standardised environment to run Plutus applications, with disciplined state man-
agement, discoverable interfaces by external clients (primarily wallets), track on-
chain information that smart contract uses, and deal with requests such as running
contract instances, forwarding user input to those instances, and notifying these
instances of ledger state change events.

Figure 27: Plutus Application Back-end schematic.

. 72

Portfolio Manager (PM) An experienced professional responsible for making and car-
rying out investment decisions on behalf of vested individuals or institutions. Clients
either invest directly into portfolio manager’s investment strategies[27, 26, 29], or
vest their funds with the investment institutions, that has multiple portfolio man-
agers working for it, and managing investors’ money.. 72

Profit & Loss (P&L, PnL) A financial statement summarising the revenues, costs,
expenses, and losses of a given period. In trading realized and unrealized P&L is
distinguished, realized P&L refers to closed positions, where unrealized P&L refers
to the current profits and losses on the open positions (e.g., not matured options)..
72

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 80 of 88

Quantitative Finance (QuantFi) Statistical branch of mathematics concerned with

modelling financial markets for the purpose of investment management. Quantita-
tive finance enables modelling of complex stochastic (random processes) and build-
ing efficient models and accurate predictions. QuantFi is used heavily in TradFi in
modelling markets, managing risk, making both automated and manual investment
decisions, to name a few.. 72

Sharpe ratio Reward to risk (variability) ratio measuring the performance of an in-
vestment as compared to a risk-free asset, after adjusting for the risk (defined as
standard deviation of the investment, i.e. volatility). The formula is named after
William Sharpe, its inventor, and defined as Sa = E[Rσaa−Rb ] ., where Ra is the asset
return, Rb is risk-free return, E[Ra − Rb ] is the expected excess return, and σa is
the standard deviation of the asset excess return. 16, 62, 76
slippage The difference between the expected price of a trade and the price at which the
trade is executed. Slippage occurs most frequently in TradFi during the periods of
high volatility and in DeFi due to inefficiency of liquidity providing formulas. To
decrease slippage, DEXs often rely on TVL.. 12, 38, 39, 77, 80
Stake Pool Operation (SPO) Running PoS (proof of stake) blockchain validator nodes.
In the case of Cardano, it implies running block producing node connected to P2P
(peer to peer) network of Cardano nodes (at the time of writing P2P feature is
still under the development and relies on the central registry of relays) via SPO
operated relays.

SPO upon correct registration on the ledger and depending on its total delega-
tion and pledge, is going to be assigned the role of slot leader during each epoch.
Each validated slot and block produced results in SPO rewards that are then dis-
tributed between delegators and the stake pool operator according to the pool’s
fixed fee (usually 340 ADA) and variable fee. Out of the profits from producing
blocks, fixed fee + variable fee · total rewards is allocated to SPO, and the rest is
distributed among delegators, in direct proportion to delegation (e.g., if delegator
provides 1% of all delegation in the pool, she will receive 1% of all remaining re-
wards after SPO receive its share).

SPOs receive rewards, in accordance to fixed fee and variable fee to cover the op-
eration costs (including the human labor), and in the case of ISPOs and mission
driven pools is a form of delegators to support specific cause and project.

Also frequently abbreviation of Stake Pool Operator, both terms basically having
the same meaning (the entity / activity of operating stake pool).. 72

Total Value Locked (TVL) Total capital amount locked into specific contract (liquid-
ity pool). For inefficient liquidity curves such as CFMM TVL is important as it
reduces the effect of slippage and volatility. However, high TVL is not only a good
metric, in the case of TVL that’s too high, the price will be resistant to update and
present a loss for market takers and opportunity for arbitrage.. 72
trading system A trading methodology, based on market data, trader’s experience, re-
search, etc., that defines a systematic approach to take as the reaction to different

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 81 of 88

market conditions. Trading systems are great in detaching emotions (which of-
ten lead to bad trades) from decision making process. System usually defines the
position that the trader should take given the current market scenario, but can
be much more precise and implemented as automated trading algorithm. Trading
using defined systems usually leads to better trader’s performance (P&L.. 74, 75

Traditional Finance (TradFi) A phrase to contrast a traditional way of providing fi-

nancial services via a centralized intermediary vs Decentralized approach of DeFi
applications. TradFi represents traditional financial service providers such as bro-
kerages, exchanges, market makers, and banks.. 72

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 82 of 88

References
[1] Simon Peyton Jones et al. Composing Contracts: An Adventure in Finan-
cial Engineering. URL: https://www.cs.tufts.edu/~nr/cs257/archive/
simon-peyton-jones/contracts.pdf. Accessed on 2021/04/26.

[2] Cryptopedia. What Are Automated Market Makers? URL: https://www.

gemini.com/cryptopedia/amm-what-are-automated-market-makers. Accessed on
2021/04/22.

[3] Vitalik Buterin. On Path Independence. URL: https://vitalik.ca/general/2017/

06/22/marketmakers.html. Accessed on 2021/04/22.

[4] Emin Gün Sirer and Phil Daian. Bancor Is Flawed. URL: https:
//hackingdistributed.com/2017/06/19/bancor-is-flawed/. Accessed on
2021/04/22.

[5] Uniswap v1 Protocol Hayden Adams et al. Novembrer 2018. URL: https://docs.
uniswap.org/protocol/V1/introduction. Accessed on 15/10/2021.

[6] Uniswap v2 Core Hayden Adams et al. March 2020. URL: https://uniswap.org/
whitepaper.pdf. Accessed on 15/10/2021.

[7] Uniswap v3 Core Hayden Adams et al. March 2021. URL: https://uniswap.org/
whitepaper-v3.pdf. Accessed on 15/10/2021.

[8] James Chen. What is an Index? URL: https://www.investopedia.com/terms/i/

index.asp. Accessed on 2021/04/22.

[9] Mirror: assets reflected on the blockchain main page. URL: https://mirror.
finance/. Accessed on 2021/04/22.

[10] Mirror Protocol: A trading and liquidity protocol whitepaper. URL: https://
mirror.one/Mirror_Protocol_Whitepaper.pdf. Accessed on 2021/04/22.

[11] Synthetix - the derivatives liquidity protocol main page. URL: https://synthetix.
io/. Accessed on 2021/04/22.

[12] Synthetix Litepaper. Version: 1.4 (March 2020). URL: https://docs.synthetix.

io/litepaper. Accessed on 2021/04/22.

[13] The Maker Team. The Dai Stablecoin System. December 2017. URL: https://
makerdao.com/whitepaper/DaiDec17WP.pdf. Accessed on 2021/04/27.

[14] The Maker Protocol: MakerDAO’s Multi-Collateral Dai (MCD) System. URL:
https://makerdao.com/en/whitepaper/. Accessed on 2021/04/27.

[15] Mirror: Reflecting Asset Value On-Chain. URL: https://docsend.com/view/

kcsm42mqiyu5t6ej. Accessed on 2021/04/27.

[16] Samuel Brooks, Anton Jurisevic, Michael Spain, Kain Warwick. Synthetix White
Paper. A decentralized payment network and stable coin v0.8. URL: https://www.
synthetix.io/uploads/synthetix_whitepaper.pdf. Accessed on 2021/04/27.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 83 of 88

[17] Synthetix Litepaper. Version: 1.4 (March 2020). URL: https://docs.synthetix.

io/litepaper. Accessed on 2021/04/27.

[18] Edward M. Miller. Risk, Uncertainty, and Divergence of Opinion. URL:

https://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.667.5934&
rep=rep1&type=pdf. Accessed on 2021/04/23.

[19] Vitalik Buterin. Token sales and shorting. URL: https://ethresear.ch/t/token-sales-

and-shorting/376. Accessed on 2021/04/23.

[20] Akhilesh Ganti. Covered Call. URL:https://www.investopedia.com/terms/c/

coveredcall.asp. Accessed on 2021/04/23.

[21] Itamar Drechsler. Qingyi (Freda) Song Drechlser. The Shorting Premium and Asset
Pricing Anomalies. URL:https://papers.ssrn.com/sol3/papers.cfm?abstract_
id=2387099

[22] Eliezer Yudkowsky. Inadequate Equilibria - Where and How Civilizations are Stuck?
URL: https://equilibriabook.com.

[23] Options Volatility & Pricing. Advanced Trading Strategies and Techniques. Sheldon
Natenberg. 2nd edition.

[24] Active Portfolio Management. A Quantitative Approach for Producing Superior Re-
turns and Controlling Risk. Richard C. Grinold, Ronald N. Khan. 2nd edition.

[25] Derivatives - Models on Models. Espen Gaarder Haug.

[26] Systematic Trading. Robert Carver.

[27] Modern Portfolio Theory and Investment Analysis. Edwin J. Elton et al. 9th edition.

[28] The xVA Challenge. Jon Gregory. 3rd edition.

[29] Econometrics. Fumio Hayashi.

[30] Principles of Professional Speculation. Victor Sperandeo.

[31] Futures Markets. Robert W. Kolb. 3rd edition.

[32] Analysis of Financial Time Series. Ruey S. Tsay. 3rd edition.

[33] Asset Price Dynamics, Volatility, and Prediction. Stephen J. Taylor.

[34] Paul Wilmott on Quantitative Finance. Paul Willmot. 2nd edition.

[35] The Alchemy of Finance. George Soros.

[36] Money Changes Everything. William N. Goetzmann.

[37] Trading Volatility, Correlation, Term Structure, and Skew. Colin Bennet.

[38] Options, Futures, and Other Derivatives. John C. Hull. 8th edition.

[39] Nonlinear Option Pricing. Julien Guyon, Pierre Henry-Labordere.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 84 of 88

[40] Modern Investment Management. An Equilibrium Approach. BOb Litterman and

the Quantitative Research Group, Goldman Sachs Asset Management.

[41] Continuous-Time Finance. Robert C. Merton. Revised edition.

[42] Pricing and Trading Interest Rate Derivatives. A Practical Guide to Swaps. J. H.
M. Darbyshire. Revised edition.

[43] Derivatives demystified. A Step-by-Step Guide to Forwards, Futures, Swaps and

Options. Andrew M. Chisholm. 2nd edition.

[44] The Volatility Surface. A Practitioner’s Guide. Jim Gatheral.

[45] The Man Who Solved The Market. How Jim Simons Launched The Quant Revolu-
tion. Gregory Zuckerman.

[46] The Complete Guide to Option Pricing Formulas. Espen Gaarder Haug. 2nd edition.

[47] Asset Pricing. John H. Cochrane. Revised edition.

[48] Modeling Derivatives in C++. Justin London

[49] Basic Stochastic Processes. Zdzislaw Brzeźniak, Tomasz Zastawniak.

[50] Stochastic Differential Equations. Bernt Oksendal.

[51] Advances in Financial Machine Learning. Marcos Lopez de Prado.

[52] Investment Science. David G. Luenberger.

[53] Fractals and Scaling in Finance. Discontinuity, Concentration, Risk. Benoit B. Man-
delbrot.

[54] Finding Alphas. A Quantitative Approach to Building Trading Strategies. Igor

Tulchinsky et al.

[55] Brownian Motion, Martingales, and Stochastic Calculus. Jean-François Le Gall.

[56] Credit Risk. Darrell Duffiee, Kenneth J. Singleton.

[57] Stochastic Volatility Modeling. Lorenzo Bergomi.

[58] Risk and Asset Allocation. Attilio Meucci.

[59] Dynamic Hedging. Managing Vanilla and Exotic Options. Nassim Taleb.

[60] Tail Risk Hedging. Creating Robust Portfolios for Volatile Markets. Vinner Bhansali.

[61] Modelling Extremal Events: for Insurance and Finance. Paul Embrechts et al.

[62] Statistical Consequences of Fat Tails. Real World Preasymptotics, Epistemology,

and Applications. Nassim Nicholas Taleb.

[63] The Price of Fixed Income Market Volatility. Antonio Mele, Yoshiki Obayashi.

[64] Options Trading. Pricing and Volatility Strategies and Techniques. Euan Sinclair.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 85 of 88

[65] Chaos. The Amazing Science of the Unpredictable. James Gleick.

[66] Nonlinear Dynamics and Chaos. With Applications to Physics, Biology, Chemistry,
and Engineering. Steven H. Strogatz. 2nd edition.

[67] Options Volatility & Pricing. Advanced Trading Strategies and Techniques. Sheldon
Natenberg. 2nd edition.

[68] Active Portfolio Management. A Quantitative Approach for Producing Superior Re-
turns and Controlling Risk. Richard C. Grinold, Ronald N. Khan. 2nd edition.

[69] Derivatives - Models on Models. Espen Gaarder Haug.

[70] Systematic Trading. Robert Carver.

[71] Modern Portfolio Theory and Investment Analysis. Edwin J. Elton et al. 9th edition.

[72] The xVA Challenge. Jon Gregory. 3rd edition.

[73] Econometrics. Fumio Hayashi.

[74] Principles of Professional Speculation. Victor Sperandeo.

[75] Futures Markets. Robert W. Kolb. 3rd edition.

[76] Analysis of Financial Time Series. Ruey S. Tsay. 3rd edition.

[77] Asset Price Dynamics, Volatility, and Prediction. Stephen J. Taylor.

[78] Paul Wilmott on Quantitative Finance. Paul Willmot. 2nd edition.

[79] The Alchemy of FInance. George Soros.

[80] Money Changes Everything. William N. Goetzmann.

[81] Trading Volatility, Correlation, Term Structure, and Skew. Colin Bennet.

[82] Options, Futures, and Other Derivatives. John C. Hull. 8th edition.

[83] Nonlinear Option Pricing. Julien Guyon, Pierre Henry-Labordere.

[84] Modern Investment Management. An Equilibrium Approach. BOb Litterman and

the Quantitative Research Group, Goldman Sachs Asset Management.

[85] Continuous-Time Finance. Robert C. Merton. Revised edition.

[86] Pricing and Trading Interest Rate Derivatives. A Practical Guide to Swaps. J. H.
M. Darbyshire. Revised edition.

[87] Derivatives demystified. A Step-by-Step Guide to Forwards, Futures, Swaps and

Options. Andrew M. Chisholm. 2nd edition.

[88] The Volatility Surface. A Practitioner’s Guide. Jim Gatheral.

[89] The Man Who Solved The Market. How Jim Simons Launched The Quant Revolu-
tion. Gregory Zuckerman.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 86 of 88

[90] The Complete Guide to Option Pricing Formulas. Espen Gaarder Haug. 2nd edition.

[91] Asset Pricing. John H. Cochrane. Revised edition.

[92] Modeling Derivatives in C++. Justin London

[93] Basic Stochastic Processes. Zdzislaw Brzeźniak, Tomasz Zastawniak.

[94] Stochastic Differential Equations. Bernt Oksendal.

[95] Advances in Financial Machine Learning. Marcos Lopez de Prado.

[96] Investment Science. David G. Luenberger.

[97] Fractals and Scaling in Finance. Discontinuity, Concentration, Risk. Benoit B. Man-
delbrot.

[98] Finding Alphas. A Quantitative Approach to Building Trading Strategies. Igor

Tulchinsky et al.

[99] Brownian Motion, Martingales, and Stochastic Calculus. Jean-François Le Gall.

[100] Credit Risk. Darrell Duffiee, Kenneth J. Singleton.

[101] Stochastic Volatility Modeling. Lorenzo Bergomi.

[102] Risk and Asset Allocation. Attilio Meucci.

[103] Dynamic Hedging. Managing Vanilla and Exotic Options. Nassim Taleb.

[104] Tail Risk Hedging. Creating Robust Portfolios for Volatile Markets. Vinner
Bhansali.

[105] Modelling Extremal Events: for Insurance and Finance. Paul Embrechts et al.

[106] Statistical Consequences of Fat Tails. Real World Preasymptotics, Epistemology,

and Applications. Nassim Nicholas Taleb.

[107] The Price of Fixed Income Market Volatility. Antonio Mele, Yoshiki Obayashi.

[108] Options Trading. Pricing and Volatility Strategies and Techniques. Euan Sinclair.

[109] Chaos. The Amazing Science of the Unpredictable. James Gleick.

[110] Nonlinear Dynamics and Chaos. With Applications to Physics, Biology, Chemistry,
and Engineering. Steven H. Strogatz. 2nd edition.

[111] Vitalik Buterin. ”Let’s run on-chain decentralized exchanges the way we run pre-
diction markets” Reddit post from 2016/10/03. (The consideration of more tradi-
tional buy-side / sell-side order book model and the inception of the famous x * y
= k AMM model). URL: https://www.reddit.com/r/ethereum/comments/55m04x/
lets_run_onchain_decentralized_exchanges_the_way/. Accessed on 2021/04/22.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 87 of 88

[112] Nick Johnson. Euler: The simplest exchange and currency. (Euler: en pricing model
for providing tokens from LP. Initial price determination via distributed farm offer-
ing). URL: https://www.reddit.com/r/ethereum/comments/54l32y/euler_the_
simplest_exchange_and_currency/. Accessed on 2021/04/22.

[113] Gemini market making model. (buy/sell order submission in the time-window,
matching at the end of the time window, more efficient price discovery via auction sys-
tem). URL: https://www.gemini.com/fees/marketplace. Accessed on 2021/04/22.

[114] Revenue equivalence. URL: https://en.wikipedia.org/wiki/Revenue_

equivalence. Accessed on 2021/04/21.

[115] Vitalik Buterin. Improving front running resistance of

x*y=k market makers. URL: https://ethresear.ch/t/
improving-front-running-resistance-of-x-y-k-market-makers/1281. Ac-
cessed on 2021/04/21.

[116] Decentralized exchanges research. URL: https://ethresear.ch/c/

decentralized-exchanges/17. Accessed on 2021/04/21.

[117] A Practical Liquidity-Sensitive Automated Market Maker. Abraham Othman, et al.

URL: https://www.cs.cmu.edu/~sandholm/liquidity-sensitive%20automated%
20market%20maker.teac.pdf. Accessed on 2021/04/21.

[118] Nassim Nicholas Taleb. The Black Swan. The Impact of the Highly Improbable.

[119] George Soros. The Alchemy of Finance: The New Paradigm.

[120] Deep Thinking. Where Artificial Intelligence Ends. . . And Human Creativity Be-
gins. Garry Kasparov.

[121] Antifragile. Things that Gain from Disorder. Nassim Nicholas Taleb.

[122] Quasi Rational Economics. Richard H. Thaler.

[123] behavioral Finance. Understanding the Social, Cognitive, and Economic Debates.
Edwin T. Burton, Sunit N. Snah.

[124] behavioral Investing. A Practitioner’s Guide to Applying behavioral Finance. James

Montier.

[125] Advances in behavioral Finance. Richard H. Thaler et al.

[126] Game Theory. An Introduction. Steven Tadelis.

[127] Ultra Society. Peter Turchin.

[128] Predictably Irrational. The Hidden Forces that Shape Our Decisions. Dan Ariely.

[129] Cardano fee structure. URL: https://docs.cardano.org/explore-cardano/

fee-structure. Accessed on 15/10/2021.

[130] The OODA (observe–orient–decide–act) loop. URL: https://en.wikipedia.org/

wiki/OODA_loop Accessed on 15/10/2021.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021

Maladex: Research-Driven Cardano DEX 88 of 88

[131] Sovan Mitra and Tong Ji. Risk measures in quantitative finance. Int. J. Business
Continuity and Risk Management, Vol. 1, No. 2, 2010.

[132] Understanding StableSwap (Curve). URL: https://miguelmota.com/blog/

understanding-stableswap-curve/. Accessed on 15/10/2021.

[133] Improved Price Oracles: Constant Function Market Makers Guillermo Angeris,
Tarun Chitra. June 2020. URL: https://web.stanford.edu/~guillean/papers/
constant_function_amms.pdf.

[134] Homogeneous Properties of Automated Market Makers. Johannes Jensen et al.

URL: https://arxiv.org/pdf/2105.02782.pdf.

[135] Park, Andreas. The Conceptual Flaws of Constant Product Automated Market
Making. August 18, 2021. URL: https://ssrn.com/abstract=3805750orhttp://
dx.doi.org/10.2139/ssrn.3805750

[136] Hydra: Fast Isomorphic State Channels Manuel M. T. Chakravarty et al. URL:
https://eprint.iacr.org/2020/299.pdf.

[137] Hydra PoC Source Code on GitHub URL: https://github.com/

input-output-hk/hydra-poc. Accessed on 15/10/2021.

[138] Ethereum - Top Contracts and Projects by Gas, Transac-

tions, and Users. @msilb7. URL: https://dune.xyz/msilb7/
Ethereum-Top-Contracts-and-Projects-by-Usage. Accessed on 21/10/2021.

[139] Uniswap: exceeded the 100,000 transactions per day. Marco

Cavicchioli. URL: https://en.cryptonomist.ch/2020/08/10/
uniswap-100000-transactions-per-day/. Accessed on 21/10/2021.

[140] PancakeSwap Continues Cook Rivals as Daily Transactions

Close on 2M. Martin Young. URL: https://cryptopotato.com/
pancakeswap-continues-cook-rivals-as-daily-transactions-close-on-2m/
Accessed on 21/10/2021.

[141] Binance in 2021: Innovating in an Increasingly Decentralized

World URL: https://www.binance.com/en/blog/421499824684901411/
binance-in-2021-innovating-in-an-increasingly-decentralized-world
Accessed on 21/10/2021.

[142] Nasdaq Daily Market Summary. URL: http://www.nasdaqtrader.com/Trader.

aspx?id=DailyMarketSummary. Accessed on 21/10/2021.

Maladex – Research-Driven Cardano DEX Date: 21 October 2021