MA 271: Financial Engineering - I
Lecture 21
Prof. Siddhartha Pratim Chakrabarty
Department of Mathematics
Indian Institute of Technology Guwahati
Time Value of Options
1 We begin with a terminology used often.
2 We say that at time t, a call option, with strike price X is:
(A) In-the-money if S(t) > X .
(B) At-the-money if S(t) = X .
(C) Out-of-the-money if S(t) < X .
3 We say that at time t, a put option, with strike price X is:
(A) In-the-money if S(t) < X .
(B) At-the-money if S(t) = X .
(C) Out-of-the-money if S(t) > X .
4 A somewhat less precise terminologies sometimes used are
deep-in-the-money and deep-out-of-the-money, to mean that the
difference between the two sides in the respective inequalities is significant.
5 An American option, which is in-the-money will bring a positive payoff, if
exercised immediately.
6 But in case of European options, this is not the case, because an European
option, which is in-the-money is nothing more than a promising asset.
Definition 1
At time t ≤ T , the intrinsic value of a call (put) option with strike price X is
(S(t) − X )+ ((X − S(t))+ ).
Definition 2
The time value of an option is the difference between the price of the option
and its intrinsic value. Accordingly, the time value is given by:
(A) European call: CE (t) − (S(t) − X )+ .
(B) European put: PE (t) − (X − S(t))+ .
(C) American call: CA (t) − (S(t) − X )+ .
(D) American put: PA (t) − (X − S(t))+ .
Definition 3
An exotic option is one in which the final payoff is path dependent i.e., it
depends on the value of the underlying asset at various time points before the
final payoff.
Barrier Option
1 A barrier option is an option whose payoff is switching in nature and
depends on whether the underlying asset prices cross a pre-defined
threshold level during the lifetime of the option.
2 A down-and-out barrier call option has the payoff for a European call
option provided the asset price does not go below a pre-specified barrier
B < S(0) and zero if it does.
3 A down-and-in barrier call option has the payoff for a European call option
provided the asset price goes below a pre-specified barrier B < S(0) and
zero if it does not.
4 An up-and-out barrier call option has the payoff for a European call option
provided the asset price does not go above a pre-specified barrier
B > S(0) and zero if it does.
5 An up-and-in barrier call option has the payoff for a European call option
provided the asset price goes above a pre-specified barrier B > S(0) and
zero if it does not.
6 The payoffs for barrier put options are similar.
Lookback Option
1 A lookback option is one whose payoff depends on either the maximum or
the minimum price of the underlying asset during the lifetime of the
option.
2 Fixed strike lookback call and put option have payoffs max (Smax − X , 0)
and max (X − Smin , 0) respectively.
3 Floating strike lookback call and put option have payoffs
max (S(T ) − Smin , 0) and max (Smax − S(T ), 0) respectively.
4 Smax and Smin denote the maximum and the minimum stock prices
respectively, during the lifetime of the option.
Asian Option
1 The payoffs for Asian options are functions of the average price of the
stock during the lifetime of the option.
2 The average price Asian call and put options have payoffs
max (Savg − X , 0) and max (X − Savg , 0) respectively.
3 Similarly, the average strike Asian call option and put options have payoffs
max (S(T ) − Savg , 0) and max (Savg − S(T ), 0) respectively.
ZT
1
4 Here, Savg := S(τ )dτ , in continuous time setup and
T
0
N
1 X
Savg := S(ti ), in discrete time setup.
N i=1
Binomial Model
1 Recall that the future stock price S(t) at time t, 0 < t ≤ T is a random
variable, with the initial stock price S(0) > 0 given.
2 We consider evenly spaced discrete time points t = 0, h, 2h, . . . , nh, . . .
with the final time being T = Nh.
3 Here, h > 0 is the length of a single time-step.
4 For convenience, we will use the notation S(n) to mean S(nh).
5 The goal is to build a concrete model for the random variable S(n).
Single Step Model
1 We consider a single step model for the random variable S(1).
2 In order to define S(1), we will need a probability space.
3 Since, in a binomial model, we allow S(1) to take only two values, it is
enough to consider Ω = {u, d}, along with a field F of all subsets of Ω,
and a probability P determined by a single number p such that P(u) = p,
P(d) = 1 − p.
4 It is possible to have other alternatives for Ω.
Single Step Model (Contd ...)
5 Let,
S(1) : Ω → (0, +∞)
be given by:
S(1) = S(0)(1 + K ),
where the rate of return K is random and has the form:
(
U , if ω = u,
Kω =
D , if ω = d,
where −1 < D < U.
6 Then: (
ω S(0)(1 + U) , if ω = u,
S (1) =
S(0)(1 + D) , if ω = d.
