2.1 Course-level Learning Outcomes ..................................................................................................

2.2 Module Breakdown ............................................................................................................................. 4

3.1 Module-level Learning Outcomes ................................................................................................. 5

3.2 Transcripts and Notes ....................................................................................................................... 6

3.2.1 Notes: Modeling on a Tree ......................................................................................................... 6

3.2.2 Transcript: Looking at the Binomial Model ....................................................................... 9

3.2.3 Notes: Pricing on a Tree............................................................................................................ 11

3.2.4 Transcript: Pricing an Option in the Binomial Model ................................................. 14

3.2.5 Notes: Hedging ............................................................................................................................ 16

3.2.6 Transcript: Hedging an Option in the Binomial Model ..............................................20

3.2.7 Notes: Exotic Derivatives ........................................................................................................ 22

3.2.8 Transcript: How to Price Some Exotic Options .............................................................. 25

3.2.9 Notes: Problem Set..................................................................................................................... 27

3.3 Additional Resources ....................................................................................................................... 31

This document contains the core content for Module 5 of Discrete-time Stochastic
Processes, entitled The Binomial Model. It consists of four sets of notes, four video
lecture transcripts, and a problem set.

Discrete-time Stochastic Processes is the third course present in the WorldQuant

University (WQU) Master of Science in Financial Engineering (MScFE) program. The
course introduces derivative pricing in discrete time. It begins with measure-theoretic
probability and stochastic processes, with an emphasis on discrete-time martingales.
These ideas are then applied to the pricing of derivatives in discrete time, followed by
an introduction to interest rate and credit risk modeling.
After completing the Discrete-time Stochastic Processes course, you will be able to::

1 Understand the language of measure-theoretic probability.

2 Understand stochastic processes and their applications.
3 Understand the theory of discrete-time martingales.
4 Define trading strategies in discrete time.
5 Create replicating portfolios in discrete time.
6 Model stock price movements on a binomial tree.
7 Price and hedge European derivatives in discrete time.
8 Price and hedge exotic European derivatives in discrete time.
9 Price and hedge American derivatives on a binomial tree.
10 Construct a simple interest rate model on a tree.
11 Price interest rate derivatives on a tree.
The Discrete-time Stochastic Processes course consists of the following one-week
modules:

1 Probability Theory
2 Stochastic Processes
3 Discrete Martingales
4 Trading in Discrete Time
5 The Binomial Model
6 American Derivatives
7 An Introduction to Interest Rate Models
This module is about a popular discrete-time model for asset prices – the binomial
model – which is used to represent asset dynamics for discrete processes, as the asset
values change randomly. The module begins by defining the binomial model in detail
and then proceeds to price different kinds of derivatives in this model.

After completing this module, you will be able to:

1 Construct a binomial tree.

2 Price and hedge vanilla European derivatives on a tree.
3 Price and hedge path-dependent and exotic European options on a tree.
We describe a special market ((𝛺, ℱ, 𝔽, ℙ), 𝑋) with trading occurring at times 𝑡 = 0,1, … , 𝑇,
where 𝑇 is a positive integer. First, we will assume that the market has only one risky
asset, 𝑋 (i.e. 𝑑 = 1). Thus, the primary tradeable assets can be represented as a vector
(1, 𝑋) where 1 is the risk-free asset.

The binomial model or binomial tree assumes a very simple form: at each time 𝑡, the
price of 𝑋 at the next time step (𝑡 + 1) can take only two values, both of which are
multiples of 𝑋𝑡 . To be precise, let 𝑢 and 𝑑 be positive real numbers with 𝑑 < 𝑢. We will
assume that for every 𝑡 ∈ {1, … 𝑇},

𝑋𝑡 = 𝑋𝑡−1 𝑢 𝑍𝑡 𝑑 (1−𝑍𝑡 ) ,

where the 𝑍𝑡 's are i.i.d Bernoulli random variables. Thus,

𝑢𝑋 if 𝑍𝑡 = 1
𝑋𝑡 = { 𝑡−1
𝑑𝑋𝑡−1 if 𝑍𝑡 = 0.

We can interpret this as follows: if we know the value of 𝑋𝑡−1 , the value of 𝑋𝑡 can either
jump “up” to 𝑢𝑋𝑡−1 or jump “down” to 𝑑𝑋𝑡−1 . We have not yet assigned probabilities to
these jumps.

If we proceed inductively, we can write (for each 𝑡 ≥ 1):

 𝑡 𝑡
𝑢 ∑𝑖=1 𝑍𝑖 𝑡 𝑢 𝑌𝑡
𝑋𝑡 = 𝑋0 ∏ 𝑢 𝑍𝑖 𝑑(1−𝑍𝑖) = 𝑋0 ( ) 𝑑 = 𝑋0 ( ) 𝑑𝑡 ,
𝑑 𝑑
𝑖=1

where 𝑌𝑡 ≔ ∑𝑡𝑖=1 𝑍𝑖 has a binomial distribution.

With these heuristics, we can now formally define all the components of the market
((𝛺, ℱ, 𝔽, ℙ), 𝑋). First, let

Ω ≔ {𝜔 = (𝜔1 , … , 𝜔 𝑇 ): 𝜔𝑖 ∈ {0, 1}}.

Define ℱ ≔ 2𝛺 and assume that ℙ({𝜔}) > 0 for each 𝜔 ∈ 𝛺. Define the sequence
{𝑍𝑡 : 𝑡 = 1, … , 𝑇} as

𝑍𝑡 (𝜔) ≔ 𝜔𝑡 , 𝜔 ∈ Ω, 1 ≤ 𝑡 ≤ 𝑇.

Define the filtration 𝔽 = (ℱ𝑡 )𝑡≥0 by:

ℱ0 ≔ {∅, 𝛺}, and ℱ𝑡 ≔ 𝜎({𝑍1 , … , 𝑍𝑡 }) 𝑡 ≥ 1.

We can then define the price process 𝑋 = {𝑋𝑡 : 𝑡 = 0.1, … , 𝑇} as

𝑡 𝑡
𝑢 ∑𝑖=1 𝑍𝑖 𝑡
𝑋0 = constant, and 𝑋𝑡 = 𝑋0 ∏ 𝑢 𝑍𝑖 𝑑(1−𝑍𝑖) = 𝑋0 ( ) 𝑑 , 𝑡 ≥ 1,
𝑑
𝑖=1

where 0 < 𝑑 < 𝑢 are fixed constants.

We now consider martingale measures on this market. If ℙ∗ is a martingale measure,

then for every 𝑡 ≥ 1,

𝑋𝑡−1 = 𝔼∗ (𝑋𝑡 |ℱ𝑡−1 ) = 𝑋𝑡−1 𝔼∗ (𝑢 𝑍𝑡 𝑑(1−𝑍𝑡 ) |ℱ𝑡−1 ),

which implies that

 𝔼∗ (𝑢 𝑍𝑡 𝑑(1−𝑍𝑡 ) |ℱ𝑡−1 ) = 1.

If we let 𝑝∗ (𝑍1 , … , 𝑍𝑡−1 ) ≔ ℙ(𝑍𝑡 = 1|ℱ𝑡−1 ), then

up* (𝑍1 , … , 𝑍𝑡−1 ) + (1 − 𝑝∗ (𝑍1 , … , 𝑍𝑡−1 ))𝑑 = 1,

which implies that

1−𝑑
𝑝∗ (𝑍1 , … , 𝑍𝑡−1 ) = 𝑢−𝑑 =: 𝑝∗ .

It is left as an exercise to show that this implies that the random variables 𝑍𝑡 are i.i.d
Bernoulli random variables with parameter 𝑝∗ , provided 𝑑 < 1 < 𝑢. Thus, the market has
a unique EMM if and only if 𝑑 < 1 < 𝑢. We shall henceforth make this assumption.
Hi, in this video, we study a simple but important example of a complete market model
called the binomial model.

In this model we will assume that the market consists of only one risky asset, 𝑋, and
the riskless asset 1.

The risky asset is assumed to evolve as follows: 𝑋0 is fixed, and for every 𝑡 ∈
{0, 1, 2, … , 𝑇 − 1}, 𝑋𝑡+1 is a function of 𝑋𝑡 :

𝑢𝑋
𝑋𝑡+1 = { 𝑡 ,
𝑑𝑋𝑡

where 0 < 𝑑 < 𝑢.

This model has a unique EMM if and only if 𝑑 < 1 < 𝑢 and, in this case, the risk neutral
conditional probability of an upward movement is:

1−𝑑
𝑝∗ = .
𝑢−𝑑

Let 𝑌𝑡 be the number of up movements up to time 𝑡 (with 𝑌0 ∶= 0). Then 𝑌 has a binomial
distribution with parameters 𝑡 and 𝑝∗ . That is,

𝑡
ℙ(𝑌𝑡 = 𝑦) = (𝑦) 𝑝∗𝑦 (1 − 𝑝∗ )𝑡−𝑦 𝑦 = 0, 1, … , 𝑡.

The random variables 𝑍𝑡 ≔ 𝑌𝑡 − 𝑌𝑡−1 are i.i.d Bernoulli random variables.

If 𝐻 is a contingent claim, then the price of 𝐻 is given by:

 𝑇
∗ (𝐻)
𝜋(𝐻) = 𝔼 = ∑ 𝐻(𝑧) ∏ 𝑝∗𝑧𝑡 (1 − 𝑝∗ )1−𝑧𝑡 ,
𝑧∈𝑍 𝑡=1

where 𝑍 = {𝑧 = (𝑧1 , … , 𝑧𝑇 ) ∶ 𝑧𝑡 ∈ {0, 1}}.

If 𝐻 is a derivative that depends only on the terminal value of 𝑋, 𝐻 = ℎ(𝑋𝑇 ), then we can
write the price as:

𝑇
∗ (𝐻) ∗ 𝑌𝑇 𝑇−𝑌𝑇 )) 𝑇
𝜋(𝐻) = 𝔼 = 𝔼 (ℎ(𝑋0 𝑢 𝑑 = ∑ ℎ(𝑋0 𝑢𝑌𝑇 𝑑𝑇−𝑌𝑇 ) ( ) 𝑝∗𝑦 (1 − 𝑝∗ )𝑇−𝑦 .
𝑦
𝑦=0
We now want to price contingent claims in the market constructed in the previous
section.

Let 𝐻 be a contingent claim. Since the market is complete, the unique no-arbitrage
price of 𝐻 is given by 𝜋(𝐻) = 𝔼∗ (𝐻). From the previous section, we can write ℙ∗ as:

ℙ∗ ({𝜔}) = ∏𝑇𝑡=1 𝑝∗ 𝜔𝑡 (1 − 𝑝∗ )1−𝜔𝑡 , 𝜔 = (𝜔1 , … , 𝜔 𝑇 ) ∈ Ω.

Hence,
𝑇
∗ (𝐻)
𝜋(𝐻) = 𝔼 = ∑ 𝐻(𝜔) ∏ 𝑝∗𝜔𝑡 (1 − 𝑝∗ )1−𝜔𝑡 .
𝜔𝜖𝛺 𝑡=1

If 𝐻 is a derivative that depends only on the terminal value of 𝑋, 𝐻 = ℎ(𝑋𝑇 ), then we can
write the price as

𝑇
∗ (𝐻) ∗ 𝑌𝑇 𝑇−𝑌𝑇 )) 𝑇
𝜋(𝐻) = 𝔼 = 𝔼 (ℎ(𝑋0 𝑢 𝑑 = ∑ ℎ (𝑋0 𝑢 𝑦 𝑑𝑇−𝑦 ) ( ) 𝑝∗ (1 − 𝑝∗ )𝑇−𝑦 ,
𝑦
𝑦=0

where 𝑌𝑇 ≔ ∑𝑇𝑖=1 𝑍𝑖 has a binomial distribution with parameters 𝑛 = 𝑇 and 𝑝 = 𝑝∗ .

Let us look at a concrete example of a call option. A call option is a derivative 𝐻 with
payoff function 𝐻 = ℎ(𝑋𝑇 ) = (𝑋𝑇 − 𝐾)+, where 𝐾 is the strike price. The price is therefore
given by:

𝑇
∗ (𝐻) 𝑇
𝜋𝐶 ≔ 𝜋(𝐻) = 𝔼 = ∑(𝑋0 𝑢 𝑦 𝑑𝑇−𝑦 − 𝐾)+ ( ) 𝑝∗𝑦 (1 − 𝑝∗ )𝑇−𝑦 .
𝑦
𝑦=0

With the following values 𝑋0 = 100, 𝐾 = 110, 𝑢 = 1/𝑑 = 1.2, 𝑇 = 2, we get

 1 − 𝑑 45 5
𝑝∗ = = =
𝑢 − 𝑑 99 11

and the price is

6 2 5 6 5 2
𝜋𝐶 = (100 × 1.2−2 − 110)+ ( ) + 2(100 − 110)+ ( ) ( ) + (100 × 1.22 − 110)+ ( )
11 11 11 11
5 2 850
= 34 (11) = 121
.

Another popular example is the put option with a payoff function of 𝐻 = (𝐾 − 𝑋𝑇 )+ ,

where 𝐾 is again called the strike price. The price is given by

𝑇
∗ (𝐻) 𝑇
𝜋𝑃 ≔ 𝜋(𝐻) = 𝔼 = ∑(𝐾 − 𝑋0 𝑢 𝑦 𝑑𝑇−𝑦 )+ ( ) 𝑝∗𝑦 (1 − 𝑝∗ )𝑇−𝑦 .
𝑦
𝑦=0

Using the same parameters as above (𝑋0 = 100, 𝐾 = 110, 𝑢 = 1/𝑑 = 1.2, 𝑇 = 2), we get:

6 2 5 6 5 2
𝜋𝑃 = (110 − 100 × 1.2−2 )+ ( +
) + 2(110 − 100) ( ) ( ) + (110 − 100 × 1.2 2 )+
( )
11 11 11 11
850
= 10 + .
121

Now, consider the derivative 𝐻 which is the difference between the call option and the
put option. Then the payoff of 𝐻 is:

𝑋 − 𝐾 if 𝑋𝑇 > 𝐾
𝐻 = (𝑋𝑇 − 𝐾)+ − (𝐾 − 𝑋𝑇 )+ = { 𝑇 = 𝑋𝑇 − 𝐾.
𝑋𝑇 − 𝐾 if 𝑋𝑇 ≤ 𝐾

Hence,

𝔼∗ ((𝑋𝑇 − 𝐾)+ − (𝐾 − 𝑋𝑇 )+ ) = 𝔼∗ (𝑋𝑇 − 𝐾),

which implies that

𝜋𝐶 − 𝜋𝑃 = 𝑋0 − 𝐾.

This relationship between the call price 𝜋𝐶 and the put price 𝜋𝑃 is called the put-call
parity. We can also verify for the prices calculated above as

850 850
− (10 + ) = −10 = 100 − 110.
121 121
Hi, in this video we look at an example of pricing an option in the binomial model.

Let us look at a concrete example of a call option. A call option is a derivative 𝐻 with
payoff function 𝐻 = ℎ(𝑋𝑇 ) = (𝑋𝑇 − 𝐾)+, where 𝐾 is the strike price. The price is therefore
given by:

𝑇
∗ (𝐻) 𝑇
𝜋𝐶 ≔ 𝜋(𝐻) = 𝔼 = ∑(𝑋0 𝑢 𝑦 𝑑𝑇−𝑦 − 𝐾)+ ( ) 𝑝∗𝑦 (1 − 𝑝∗ )𝑇−𝑦 .
𝑦
𝑦=0

With the following values 𝑋0 = 100, 𝐾 = 110, 𝑢 = 1/𝑑 = 1.2, 𝑇 = 2, we get:

1 − 𝑑 45 5
𝑝∗ = = = ,
𝑢 − 𝑑 99 11

and the price is

6 2 5 6 5 2
𝜋𝐶 = (100 × 1.2−2 − 110)+ ( ) + 2(100 − 110)+ ( ) ( ) + (100 × 1.22 − 110)+ ( )
11 11 11 11
5 2 850
= 34 ( ) = .
11 121

Another popular example is the put option with a payoff function of 𝐻 = (𝐾 − 𝑋𝑇 )+ ,

where 𝐾 is again called the strike price. The price is given by:

𝑇
∗ (𝐻) 𝑇
𝜋𝑃 ∶= 𝜋(𝐻) = 𝔼 = ∑(𝐾 − 𝑋0 𝑢 𝑦 𝑑𝑇−𝑦 )+ ( ) 𝑝∗𝑦 (1 − 𝑝∗ )𝑇−𝑦 .
𝑦
𝑦=0

Using the same parameters as above (𝑋0 = 100, 𝐾 = 110, 𝑢 = 1/𝑑 = 1.2, 𝑇 = 2), we get:
 6 2 5 6 5 2
𝜋𝑃 = (110 − 100 × 1.2−2 )+ ( +
) + 2(110 − 100) ( ) ( ) + (110 − 100 × 1.2 2 )+
( )
11 11 11 11
850
= 10 + .
121

Now consider the derivative 𝐻 which is the difference between the call option and the
put option. Then the payoff of 𝐻 is:

𝑋 −𝐾 if 𝑋𝑇 > 𝐾
𝐻 = (𝑋𝑇 − 𝐾)+ − (𝐾 − 𝑋𝑇 )+ = { 𝑇 = 𝑋𝑇 − 𝐾.
𝑋𝑇 − 𝐾 if 𝑋𝑇 ≤ 𝐾

Hence,

𝔼∗ ((𝑋𝑇 − 𝐾)+ − (𝐾 − 𝑋𝑇 )+ ) = 𝔼∗ (𝑋𝑇 − 𝐾),

which implies that

𝜋𝐶 − 𝜋𝑃 = 𝑋0 − 𝐾.

This relationship between the call price 𝜋𝐶 and the put price 𝜋𝑃 is called the put-call
parity. We can also verify for the prices calculated above as:

850 850
− (10 + ) = −10 = 100 − 110.
121 121
We now look at the problem of replication in a binomial tree.

Since the market is complete, we know that the unique EMM ℙ∗ has PRP, in the sense
that for every (𝔽, ℙ∗ )-martingale 𝑀, there exists a predictable process, 𝜑𝑀 , such that for
every 1 ≤ 𝑡 ≤ 𝑇,
𝑡

𝑀𝑡 − 𝑀0 = ∑ 𝜑𝑘𝑀 (𝑋𝑘 − 𝑋𝑘−1 ).

𝑘=1

The process 𝜑𝑀 advertized above can be found as follows. First note that for each 𝑡 ≥ 1,

𝑀𝑡 = 𝑀𝑡−1 + 𝜑𝑡𝑀 (𝑋𝑡 − 𝑋𝑡−1 ),

which implies that

𝑀𝑡 − 𝑀𝑡−1
𝜑𝑡𝑀 =
𝑋𝑡 − 𝑋𝑡−1

since 𝑋𝑡 ≠ 𝑋𝑡−1 .

Now, let 𝐻 be a contingent claim. Define the martingale 𝑉 𝐻 = (𝑉𝑡𝐻 : 0 ≤ 𝑡 ≤ 𝑇) by:

𝑉𝑡𝐻 ≔ 𝔼∗ (𝐻|ℱ𝑡 ), 0 ≤ 𝑡 ≤ 𝑇.

Note that 𝑉0𝐻 = 𝔼∗ (𝐻) = 𝜋(𝐻).

Now since 𝑉 𝐻 is an (𝔽, ℙ∗ )-martingale, it follows that we can find a predictable process
𝜑𝐻 such that
 𝑡

𝑉𝑡𝐻 − 𝑉0𝐻 = ∑ 𝜑𝑘𝐻 (𝑋𝑘 − 𝑋𝑘−1 ),

𝑘=1

which implies that

𝑉𝑡𝐻 = 𝜋(𝐻) + ∑ 𝜑𝑘𝐻 (𝑋𝑘 − 𝑋𝑘−1 ).

𝑘=1

Furthermore,

𝐻
𝑉𝑡𝐻 − 𝑉𝑡−1
𝜑𝑡𝐻 = .
𝑋𝑡 − 𝑋𝑡−1

Notice that since

𝑉𝑡𝐻 = 𝜋(𝐻) + ∑ 𝜑𝑘𝐻 (𝑋𝑘 − 𝑋𝑘−1 ),

𝑘=1

it follows that 𝑉 𝐻 is the value of the replicating strategy 𝜑𝐻 , and 𝑉𝑇𝐻 = 𝔼∗ (𝐻|ℱ𝑇 ) = 𝐻.

Let us look at an example. Consider the example discussed in the previous section with
the following values: 𝑋0 = 100, 𝑢 = 1/𝑑 = 1.2, 𝑇 = 2. Consider a call option with strike
price 𝐾 = 110. We want to calculate 𝜑𝐻 , where 𝐻 = (𝑋𝑇 − 110)+ .

Since 𝑇 = 2, the sample space is:

Ω = {(1, 1), (1, 0), (0, 1), (0, 0)}.

For convenience, we will replace 1 with 𝑢 and 0 with 𝑑 and write

Ω = {(𝑢, 𝑢), (𝑢, 𝑑), (𝑑, 𝑢), (𝑑, 𝑑)}.

The values of 𝑉 can be summarized in the following table:

The hedging strategy 𝜑 𝐻 is given by:

𝐻
𝑉𝑡𝐻 − 𝑉𝑡−1
𝜑𝑡𝐻 = .
𝑋𝑡 − 𝑋𝑡−1

The values are summarized below:

The interpretation of the strategy is as follows. At time 0, starting with an initial capital
850 51 −4250
of 𝑉0𝐻 = 121, buy 𝜑1𝐻 = 121 units of 𝑋 and invest the remainder 𝜂1𝐻 = 𝑉0𝐻 − 𝜑1𝐻 𝑋0 = 121
. So,
4250
the transaction involves borrowing 121
from the bank. Hold these quantities until time

1.

At time 1, if the share goes up from 100 to 120, then the value of the portfolio will be
−4250 51 170
𝑉1𝐻 ((𝑢, 𝑢)) = 𝑉1𝐻 ((𝑢, 𝑑)) = 121
+ 121 × 120 = 11
, and, in that case, we will change
 17
our holding in 𝑋 to be 𝜑2𝐻 ((𝑢, 𝑢)) = 𝜑2𝐻 ((𝑢, 𝑑)) = and the bank account investment to
22
170 17 −850
𝜂2𝐻 = 11
− 22 × 120 = 11
.

Likewise, if the stock falls to 1.2−1 × 100, the value of the portfolio will be 𝑉1𝐻 ((𝑑, 𝑢)) =
−4250 51
𝑉1𝐻 ((𝑑, 𝑑)) = + × (1.2−1 × 100) = 0, and there will be no investment in the stock
121 121

(𝜑2𝐻 ((𝑑, 𝑢)) = 𝜑2𝐻 ((𝑑, 𝑑)) = 0), and no investment in the bank account 𝜂2 = 𝑉2 − 𝜑2𝐻 𝑋2 = 0.

At time 2, if the share goes up again (to 144), the value of the portfolio will be
−850 17
𝑉2 ((𝑢, 𝑢)) = 11
+ 22 × 144 = 34. For all other cases, 𝑉 𝐻 is equal to 0, which corresponds

to the value of the derivative.

Hi, in this video we go through an example of hedging an option in the binomial model.

Consider the example discussed in the previous section, with the following values: 𝑋0 =
100, 𝑢 = 1/𝑑 = 1.2, 𝑇 = 2. Consider a call option with strike price 𝐾 = 110. We want to
calculate 𝜑𝐻 , where 𝐻 = (𝑋𝑇 − 110)+ .

Since 𝑇 = 2, the sample space is:

𝛺 = {(1, 1), (1, 0), (0, 1), (0, 0)}.

For convenience, we will replace 1 with 𝑢 and 0 with 𝑑, and write

𝛺 = {(𝑢, 𝑢), (𝑢, 𝑑), (𝑑, 𝑢), (𝑑, 𝑑)}.

The values of 𝑉 can be summarized in the following table:

The hedging strategy 𝜑 𝐻 is given by:

𝐻
𝑉𝑡𝐻 − 𝑉𝑡−1
𝜑𝑡𝐻 = .
𝑋𝑡 − 𝑋𝑡−1

The values are summarized below:

In this section we will give examples of some common exotic derivatives and show
how to price them. We will continue with the market used in the previous section.
Recall that this market has the following parameters:

𝑇 = 2, 𝑋0 = 100, 𝑢 = 1/𝑑 = 1.2 and 𝑝∗ = 5/11.

The first example we consider is an Asian option. The payoff of an arithmetic average
Asian call option with strike price 𝐾 is

𝑇 𝑇 +
1 1
𝐻 ≔ max ( ∑ 𝑋𝑡 − 𝐾, 0) = ( ∑ 𝑋𝑡 − 𝐾 ) .
𝑇+1 𝑇+1
𝑡=0 𝑡=0

Its payoff is similar to that of a vanilla (ordinary) call option, but instead of using the
terminal value of the stock price 𝑋𝑇 , one uses the average of 𝑋 over the lifetime of the
option. The are variations on the time points used to calculate the average.

In our example, the payoff looks like this:

The price can then be calculated as

34 850
𝜋(𝐻) = × 𝑝∗ 2 = .
3 363
The price of an Asian put option can be calculated in a similar way. One can also
replace the arithmetic average with a geometric average, and the corresponding option
is called a geometric average Asian option.

Next, we look at a lookback option. The payout of a lookback put option is:

𝐻 ≔ MAX 𝑋𝑡 − 𝑋𝑇 .
0≤𝑡≤𝑇

For our example, the payoff values are tabulated below:

The price of 𝐻 is calculated as follows:

5 6 275 6 2
𝜋(𝐻) = 20 × × + × ( ) ≈ 14.0496.
11 11 9 11

The last example we look at is a barrier option. A barrier option makes a payment if the
stock price hits or misses a barrier. As an example, we consider an up-and-out call
option whose payoff is

0 if 𝑚𝑎𝑥0≤𝑡≤𝑇 𝑋𝑡 ≥ 𝐵
𝐻 ∶= (𝑋𝑇 − 𝐾)+ 𝐼{𝑚𝑎𝑥0≤𝑡≤𝑇 𝑋𝑡≥𝐵} = {
(𝑋𝑇 − 𝐾)+ otherwise.
Here, 𝐾 is the strike price and 𝐵 > max(𝐾, 𝑋0 ) is the barrier. So, the option becomes void
if 𝑋 crosses the barrier 𝐵; otherwise it behaves like an ordinary call option, hence the
name “up and out”.

For parameters 𝐵 = 115 and 𝐾 = 95 we get:

The price of 𝐻 is calculated as follows:

6 5 150
𝜋(𝐻) = 5 × 11 × 11 = 121 ≈ 1.23967.
Hi, in this video, we go through examples of how to price some exotic options.

The first example we consider is an Asian option. The payoff of an arithmetic average
Asian call option with strike price 𝐾 is:

𝑇 𝑇 +
1 1
𝐻 ≔ 𝑚𝑎𝑥 ( ∑ 𝑋𝑡 − 𝐾, 0 ) = ( ∑ 𝑋𝑡 − 𝐾) .
𝑇+1 𝑇+1
𝑡=0 𝑡=0

Its payoff is similar to that of a vanilla (ordinary) call option, but instead of using the
terminal value of the stock price 𝑋𝑇 , one uses the average of 𝑋 over the lifetime of the
option. There are variations on the time points used to calculate the average.

In our example, the payoff looks like this:

The price can then be calculated as:

34 850
𝜋(𝐻) = × 𝑝∗2 = .
3 363

Next, we look at the lookback option. The payout of a lookback put option is:

𝐻 ≔ max 𝑋𝑡 − 𝑋𝑇 .
0≤𝑡≤𝑇
For our example, the payoff values are tabulated below:

The price of 𝐻 is calculated as follows:

5 6 275 6 2
𝜋(𝐻) = 20 × × + × ( ) ≈ 14.0496.
11 11 9 11
Consider a binomial tree with the following parameters:

𝑢 = 1.4, 𝑑 = 0.7, 𝑇 = 2, 𝑋0 = 100.

What is the price of a call option with strike price 𝐾 = 100 and expiring at time 𝑇 = 2?

Solution:
First of all, remember the payoff at maturity for a call option,

𝐻 ≔ max {𝑋𝑇 − 𝐾, 0}.

The payoff values are tabulated below:

The second step is always to compute the 𝑃 ∗. From the lecture notes:

1−𝑑 1 − 0.7
𝑝∗ = = ≈ 0.4286.
𝑢−𝑑 1.4 − 0.7

From the table above we know that, 𝐻({𝑢, 𝑑}) = 𝐻({𝑑, 𝑢}) = 𝐻({𝑑, 𝑑}) = 0. The price of 𝐻
is calculated as follows:
 𝜋(𝐻) = 𝔼∗ (𝐻) = 𝑃∗ ({𝑢, 𝑢})𝐻({𝑢, 𝑢}) + 𝑃∗ ({𝑢, 𝑑})𝐻({𝑢, 𝑑}) + 𝑃∗ ({𝑑, 𝑢})𝐻({𝑑, 𝑢})
+ 𝑃∗ ({𝑑, 𝑑})𝐻({𝑑, 𝑑}) = 𝑝2 × 96 ≈ 17.6327.

Consider a binomial tree with the following parameters:

𝑢 = 1.25, 𝑑 = 1/𝑢, 𝑇 = 2, 𝑋0 = 400.

Consider a hedging strategy (𝜂, 𝜑) for the derivative 𝐻 with payoff 𝐻 = 𝑋2 − 319, where
𝜑 is the investment in 𝑋. Compute the value of 𝜑1 .

Solution:
The payoff values are tabulated below:

The second step is always to compute the 𝑃 ∗. From the lecture notes:

1−𝑑 1 − 4/5 4
𝑝∗ = = = ≈ 0.444.
𝑢 − 𝑑 5/4 − 4/5 9

Notice that since

𝑉𝑡𝐻 = 𝜋(𝐻) + ∑ 𝜑𝑘𝐻 (𝑋𝑘 − 𝑋𝑘−1 ),

𝑘=1
it follows that 𝑉 𝐻 is the value of the replicating strategy 𝜑𝐻 , and 𝑉𝑇𝐻 = 𝔼∗ (𝐻|ℱ𝑇 ) = 𝐻.

For instance, we can compute 𝑉1𝐻 ({𝑢, . }) as follows:

4 5
𝑉1𝐻 ({𝑢, . }) = 𝑉1𝐻 ({𝑢, 𝑢}) = 𝑉1𝐻 ({𝑢, 𝑑}) = ∗ 306 + ∗ 81 = 181.
9 9

Following the same procedure,

4 5
𝑉1𝐻 ({𝑑, . }) = 𝑉1𝐻 ({𝑑, 𝑢}) = 𝑉1𝐻 ({𝑑, 𝑑}) = ∗ 81 − ∗ 63 = 1.
9 9

Last step is to compute 𝑉0𝐻 :

4 2 4 5 5 2
𝑉0𝐻 = ( ) ∗ 81 + 2 ∗ 81 ∗ ∗ − ( ) ∗ 63 = 81.
9 9 9 9

The values of 𝑉 can be summarized in the following table:

The hedging strategy 𝜑 𝐻 is given by

𝐻
𝑉𝑡𝐻 − 𝑉𝑡−1
𝜑𝑡𝐻 = .
𝑋𝑡 − 𝑋𝑡−1

The problem only asks about 𝜑1𝐻 , thus,

 𝑉1𝐻 − 𝑉0𝐻 181 − 81
𝜑1𝐻 = = = 1.
𝑋1 − 𝑋0 500 − 400

The solution is 𝜑1𝐻 = 1.

Follmer, H. and Schleid, A. 2002. Stochastic Finance: An Introduction in Discrete Time.
Shreve, S. 2004. Stochastic Calculus for Finance I: The Binomial Model.
Shiryaev, A. 1999. Essentials of Stochastic Finance.
BK. Risk Neutral Valuation.