Derivatives

Chapter 10 Binomial Option Pricing

1 / 44
Binomial Option Pricing: Basic Concepts
• The binomial option pricing model enables us to determine the
price of an option, given the characteristics of the stock or other
underlying asset.
• The binomial option pricing model assumes that, over a period
of time, the price of the underlying asset can move up or down
only by a specified amount—that is, the asset price follows a
binomial distribution.
• Given this assumption, it is possible to determine a no-arbitrage
price for the option.
• Surprisingly, this approach, which appears at first glance to be
overly simplistic, can be used to price options, and it conveys
much of the intuition underlying more complex (and seemingly
more realistic) option pricing models.
2 / 44
10.1 A ONE-PERIOD BINOMIAL TREE

• Computing the Option Price

• The Binomial Solution
• Arbitraging a Mispriced Option
• A Graphical Interpretation of the Binomial Formula

3 / 44
Binomial Tree

• At any point in time, the stock price can change to either an up

value or a down value. In-between, greater or lesser values are
not permitted.
• This depiction of possible stock prices is called a binomial tree.
• It is a good approximation of the reality because the price goes
up or down by 1 cent in a very short period

4 / 44
Binomial Tree

• This depiction of possible stock prices is called a binomial tree.

Example. Consider a European call option on the stock of XYZ, with

a $40 strike and 1 year to expiration. XYZ does not pay dividends
and its current price is $41. The continuously compounded risk-free
interest rate is 8%. We wish to determine the option price.

5 / 44
Computing the Option Price
Consider two portfolios
• Portfolio A: buy one call option with a $40 strike
• Portfolio B: buy 2/3 shares and borrow $18.462 at the risk-free
rate
• Note that Portfolios A and B have the same payoff
Stock Price in 1 Year
30 60

Portfolio A Payoff 0 20

2/3 purchased shares 20 40

Repay loan of 18.462 -20 (18.462×e0.08 ) -20
Portfolio B Payoff 0 20

6 / 44
Computing the Option Price

• Law of one price: Portfolios A and B should have the same cost
• Since Portfolio B costs $8.871 (2/3 × 41–18.462 = 8.871), the
price of one option must be $8.871
• Portfolio B is a synthetic call
- There is a way to create the payoff to a call by buying shares and
borrowing

7 / 44
The Binomial Solution

• How did we know that buying 2/3 of a share of stock and

borrowing $18.462 would replicate a call option?
• The problem is to solve for ∆ and B such that our portfolio of ∆
shares and B in lending duplicates the option payoff.
• Suppose that the stock has a continuous dividend yield of δ,
which is reinvested in the stock. Thus, if you buy one share at
time t, at time t+h you will have eδh shares.
• If the length of a period is h, the interest factor per period is erh .

8 / 44
The Binomial Solution
• Let S be the stock price today.
• We can write the stock price as uS when the stock goes up and
as dS when the price goes down.
• Let Cu and Cd represent the value of the option when the stock
goes up or down, respectively.
• The stock price tree and the corresponding tree for the value of
the option:

9 / 44
The Binomial Solution
• The value of the replicating portfolio at time h, with stock price
Sh , is ∆Sh eδh + erh B
Stock Price in 1 Year (S1)
Sh = dS Sh = uS

Call Option Payoff Cd Cu

∆ shares ∆dSeδh ∆uSeδh

B bonds erh B erh B
Replicating Portfolio Payoff ∆dSeδh + erh B ∆uSeδh + erh B
• The condition of a successful replication
- If Sh = dS
∆dSeδh + erh B = Cd
- If Sh = uS
∆uSeδh + erh B = Cu
10 / 44
The Binomial Solution
• This is two equations in the two unknowns ∆ and B.
• Solving for ∆ and B gives

Cu − Cd
∆ = e−δh
S(u − d)
uCd − dCu
B = e−rh
u−d
• Given the expressions ∆ for and B, we can derive a simple
formula for the value of the option.
• The cost of the option
!
−rh e(r −δ)h − d u − e(r −δ)h
C = ∆S + B = e Cu + Cd
u−d u−d

11 / 44
The Binomial Solution

Example. In the previous example, u = 60/41 = 1.4634,

d = 30/41 = 0.7317. Cu = 60 − 40 = 20. Cd = 0. δ = 0. h = 1.
• If Sh = 30
∆30 + e0.08 B = 0
• If Sh = 60
∆60 + e0.08 B = 20
• ∆ = 2/3. B = −18.462.
• The equations could be tedious, but the example is easy.

12 / 44
The Binomial Solution
• The assumed stock price movements, u and d, should not give
rise to arbitrage opportunities.

u > e(r −δ)h > d

• Suppose d < u < e(r −δ)h , we could short the stock to hold the
bonds and arbitrage.

Stock Price in 1 Year (S1)

Sh = dS Sh = uS

short e−δh shares −dS −uS

buy Se−δh bonds Se(r −δ)h Se(r −δ)h
Replicating Portfolio Payoff (e(r −δ)h − d)S (e(r −δ)h − u)S
• (e(r −δ)h − d)S > (e(r −δ)h − u)S > 0
13 / 44
Arbitraging a Mispriced Option
• What if the observed option price differs from the theoretical
price?
- buy low and sell high

Example.Suppose that the market price for the option is $9.00

instead of $8.871. We can sell the option and buy the synthetic
option, and make $0.129.

Stock Price in 1 Year

30 60

Written call 0 -20

2/3 purchased shares 20 40
Repay loan of 18.462 -20 -20
Total Payoff 0 0

14 / 44
A Graphical Interpretation of the Binomial Formula

• We choose ∆ and B to yield a portfolio that pays Cd when

Sh = dS and Cu when Sh = uS .

15 / 44
A Graphical Interpretation of the Binomial Formula

• This line runs through points E and D. This determines

∆ = (Cu − Cd )/(uS − dS).
• The point A is the value of the portfolio when Sh = 0, which is
the time-h value of the bond position erh B.
• Any line replicating a call will have a positive slope (∆ > 0) and a
negative intercept (B < 0).
• A portfolio replicating a put would have negative slope (∆ < 0)
and positive intercept (B > 0).

16 / 44
10.2 CONSTRUCTING A BINOMIAL TREE

• Continuously Compounded Returns

• Volatility
• Constructing u and d
• Estimating Historical Volatility

17 / 44
Continuously Compounded Returns

• Continuously compounded returns are mathematically

convenient and widely used in practice
• The logarithmic function computes continuously compounded
returns from prices.
- Let St and St+h be stock prices at times t and t + h. The
continuously compounded return between t and t + h is then

rt,t+h = ln(St+h /St )

• The exponential function computes prices from continuously

compounded returns.
- If we know the continuously compounded return rt,t+h , we can
obtain
St+h = St ert,t+h

18 / 44
Continuously Compounded Returns

• Continuously compounded returns are additive.

- Suppose we have continuously compounded returns over
consecutive periods—for example, rt,t+h , rt+h,t+2h , etc. The
continuously compounded return over a long period is the sum of
continuously compounded returns over the shorter periods
n
X
rt,t+nh = rt+(i−1)h,t+ih
i=1

19 / 44
Volatility
• The volatility of an asset, defined as the standard deviation of
continuously compounded returns, is a key input for any option
pricing calculation.
• The annual continuously compounded return is
rannual = 12
P
i=1 rmonthly ,i
• The variance of the annual continuously compounded return
Var (rannual ) = Var ( 12
P
i=1 rmonthly ,i )
• It is common to assume that returns are uncorrelated over time.
With this assumption, the variance of a sum is the sum of the
variances.
• Annual variance σ 2 = 12 × σmonthly2

• In general, if we split the year into n periods of length h, the

standard deviation over the period of length h
√
σh = σ h
20 / 44
Constructing u and d

• What if there were no uncertainty about the future stock price?

- The stock price next period must equal the forward price.

St+h = Ft,t+h = St e(r −δ)h

• We model the stock price evolution by adding uncertainty to the

forward price: √
uSt = Ft,t+h e+σ h
√
dSt = Ft,t+h e−σ h

• If σ = 0, uSt = dSt = Ft,t+h .

21 / 44
Constructing A Binomial Tree
• We get the stock returns from the stock prices.
• We model the stock returns u and d using the equations
√
u = e(r −δ)h+σ h

√
d = e(r −δ)h−σ h

• r is the continuously compounded annual interest rate, δ is the

continuous dividend yield, σ is the annual volatility, and h is the
length of a binomial period in years.
√
• If St+h = uSt , the stock return is (r − δ)h + σ h.
√
• If St+h = dSt , the stock return is (r − δ)h − σ h.
• We incorporate uncertainty into the stock return using volatility.
Stocks with a larger σ will have a greater chance of a return far
from (r − δ)h
22 / 44
Estimating Historical Volatility
• We measure σ by computing the standard deviation of
continuously compounded historical returns.
• Volatility computed from historical stock returns is historical
volatility.

23 / 44
One-Period Example with a Forward Tree
Example. Back to our early example. Suppose volatility is 30%.
√
Since the period is 1 year, we have h = 1, so that σ h = 0.30. We
also have S0 = 41. r = 0.08. δ = 0.
uS = 41e0.08+0.3 = 59.954
dS = 41e0.08−0.3 = 32.903

• Because the binomial tree is different, the option price will be

different as well. 24 / 44
10.3 TWO OR MORE BINOMIAL PERIODS

• A Two-Period European Call

• Many Binomial Periods

25 / 44
A Two-Period European Call

Example. After year 1, we move further up or down.

Suu = u 2 41 = 59.954e0.08+0.3 = 87.669

26 / 44
A Two-Period European Call
• How do we price the option when we have two binomial periods?

- we work backward through the binomial tree

• Year 2, Suu = 87.669, Cuu = 47.669.
• Year 2, Sud = Sdu = 48.114, Cud = Cdu = 8.114.
• Year 2, Sdd = 26.405, Cdd = 0.
• Year 1, Su = 59.954.
e(r −δ)h − d u − e(r −δ)h
Cu =e−rh (Cuu + Cud )
u−d u−d
e0.08 − 0.803 1.462 − e0.08
=e−0.08 (47.666 + 8.114 )
1.462 − 0.803 1.462 − 0.803
=23.029

• Year 1, Sd = 32.903. We could compute Cd = 3.187.

• Year 0, S = 41. We could obtain C = 10.737.
27 / 44
A Two-Period European Call

• The option price is greater for the 2-year ($10.737) than for the
1-year option ($7.839)
• The option’s ∆ and B are different at different nodes.
• ∆ increases to 1 as we go further into the money.

28 / 44
Many Binomial Periods
• Once we understand the two-period option, it is straightforward
to value an option using more than two binomial periods.

29 / 44
11.3 THE BINOMIAL TREE AND LOGNORMALITY
• The usefulness of the binomial pricing model hinges on the
binomial tree providing a reasonable representation of the stock
price distribution.
• The figure is one particular path through a 10,000-step binomial
tree.

30 / 44
11.3 THE BINOMIAL TREE AND LOGNORMALITY
• The binomial tree approximates a lognormal distribution, which
is commonly used to model stock prices.

31 / 44
10.4 PUT OPTIONS
• The binomial method easily accommodates put options.

32 / 44
10.5 AMERICAN OPTIONS

• The binomial method is well-suited to valuing American options.

• It is easy to check at each node whether early exercise is
optimal.
• If the value of the option is greater when exercised, we assign
the exercised value to the node.
• Otherwise, we assign the value of the option unexercised.
• We work backward through the tree as usual.

33 / 44
American Options
• The only difference: $30.585. The American option at that point
is worth $9.415 (40-30.585), its early-exercise value. The value
of the option if unexercised is $8.363.

34 / 44
10.6 OPTIONS ON OTHER ASSETS

• Option on a Stock Index

• Options on Currencies

35 / 44
Option on a Stock Index
• This is the same as a stock option.

36 / 44
Options on Currencies

• With a currency with spot price x0 , the forward price is

F0,h = x0 e(r −rf )h , where rf is the foreign interest rate.
• We construct the binomial tree using
√
ux = xe(r −rf )h+σ h

√
dx = xe(r −rf )h−σ h

• If we think of rf as the dividend yield on the foreign currency, the

equations look exactly like those for an index option.

37 / 44
Options on Currencies

38 / 44
Summary: Recipe for A One-period Tree
1. Collect information S, r , δ, h, K
2. Estimate the volatility σ
3. Construct a binomial tree u, d, uS, dS
√ √
u = e(r −δ)h+σ h
, d = e(r −δ)h−σ h

4. Compute the call option price Cu , Cd

5. Compute ∆, B, and the option price C

Cu − Cd uCd − dCu
∆ = e−δh , B = e−rh
S(u − d) u−d
C = ∆S + B
- If we just want the option price, use the direct formula!
e(r −δ)h − d u − e(r −δ)h
C = e−rh Cu + Cd
u−d u−d
39 / 44
Summary: Recipe for A Two-period Tree
1. Collect information S, r , δ, h, K
- h is half of the time to expiration
2. Estimate the historical volatility σ
3. Construct a binomial tree u, d, Su = uS, Sd = dS, Suu = u 2 S,
Sdd = d 2 S, Sdu = Sud = udS
√ √
u = e(r −δ)h+σ h
, d = e(r −δ)h−σ h

4. Compute the call option price Cuu , Cdu = Cud , Cdd

5. Compute Cu and Cd at node u and d
!
−rh e(r −δ)h − d u − e(r −δ)h
Cu = e Cuu + Cud
u−d u−d
!
−rh e(r −δ)h − d u − e(r −δ)h
Cd = e Cdu + Cdd
u−d u−d
6. Compute C at the initial node
40 / 44
A Case Study: HSBC Option

• Let’s apply our binomial model and price the call option of HSBC.
• Today is March 26, 2019.
• The spot price of HSBC (00005) is 63.90.
• The interest rate is 0.21%.
• Our American call option has a strike price of 65.
• The expiration date is April 28, 2019. (33 days; h = 33/365)
• No dividends are paid during this period.
• What is the option price?

41 / 44
A Case Study: HSBC Option

1. Collect information S, r , δ, h, K
- www.hkex.com.hk/Market-Data/Securities-Prices/Equities
2. Estimate the volatility σ = 14%
- We could estimate it using daily stock return data
- www.hkex.com.hk/eng/sorc/options/statistics_hv_iv.aspx
- HSBC_Volatility.xslx
3. Construct a binomial tree u, d, uS, dS
4. Compute the call option price Cu , Cd
5. Compute ∆, B, and the option price C
- Details are in the spreadsheet

42 / 44
A Case Study: HSBC Option

• We obtain the price from the binomial model

- 0.81 in a one-period tree
- 0.69 in a two-period tree
• HKEX and many trading platforms provide the solution to a large
tree
- www.hkex.com.hk/eng/sorc/tools/calculator_stock_option.aspx
- 0.64 in this case (HSBC_K_65_sigma_14.pdf)
• The actually trading price is 0.62 (HSBC_Price.pdf)
• Our binomial option pricing model is quite good

43 / 44
A Case Study: HSBC Option

• Surprisingly, this approach, which appears at first glance to be

overly simplistic, is quite useful
• The success of the binomial model is across the board
- Details are in the pdfs

Binomial Model Market Price

strike call put call put

60 3.992 0.081 4.19 0.14

62.5 1.928 0.516 1.97 0.49
65 0.644 1.732 0.62 1.57
67.5 0.135 3.725 0.13 3.59

44 / 44