Option Pricing in

Continuous Time

1
Stochastic Processes
Describe the way in which a variable such as
a stock price, exchange rate or interest rate
changes through time
Incorporates randomness reflecting the
uncertainty in the values of these market
variables in the future

2
Example 1
Each day a stock price
increases by $1 with probability 30%
stays the same with probability 50%
reduces by $1 with probability 20%

3
Example 2
Each day a stock price change is drawn from
a normal distribution with mean $0.2 and
standard deviation $1

4
Outline of lecture
Basic introduction to stochastic processes
(see FE 610)
Wiener process, aka Brownian motion
Geometric Brownian motion
Black-Scholes-Merton model
Option pricing in the BSM model
Volatility

5
Markov Processes (See pages 301-302)
Markov property: The past history is
irrelevant for the future evolution. Only
present state matters
Future movements in a variable
depend only on where we are, not the
history of how we got to where we are
We assume that stock prices follow
Markov processes

6
Weak-Form Market Efficiency
This asserts that it is impossible to
produce consistently superior returns with a
trading rule based on the past history of
stock prices. In other words technical
analysis does not work.
A Markov process for stock prices is
consistent with weak-form market
efficiency
7
Example
A variable is currently 40
It follows a Markov process
The changes of the variable over
equal non-overlapping time intervals
are i.i.d. (independent identically
distributed).
At the end of 1 year the variable has a
normal probability distribution with
mean 40 and standard deviation 10
8
Questions
What is the probability distribution
of the stock price at the end of 2
years?
½ years?
¼ years?
Dt years?

Taking the limit Dt 0 defines a

continuous time stochastic 9
Variances & Standard Deviations
In Markov processes changes in
successive periods of time are
independent
This means that variances are
additive
Standard deviations are not additive

10
A Wiener Process (See pages 302-303)
Define f(m,s2) as a normal distribution
with mean m and variance s2
A variable z follows a Wiener process
(aka Brownian motion) if:
The
z change in z in
 t where (0,1) interval of time Dt is
is any

The values of Dz for any 2 different (non-

overlapping) periods of time are
independent
11
Properties of a Wiener Process

Mean of [z (T ) – z (0)] is 0
Variance of [z (T ) – z (0)] is T
T
Standard deviation of [z (T ) – z (0)] is
The changes [z(T3)-z(T2)] and [z(T1)-z(T0)]
are independent (

12
Taking Limits . . .
The time increments can be taken
arbitrarily small in a random walk with
normally distributed increments
In the limit as Dt tends to zero we get a
continuous time stochastic process: the
Wiener process, or Brownian motion
The process is “fractal-looking”: it does
not have a derivative at any point.
13
The Example Revisited
A stock price starts at 40 and has a probability
distribution of f(40,100) at the end of the year
If we assume the stochastic process is Markov with no
drift, then the process is
dS = 10dz
If the stock price were expected to grow by $8 on
average during the year, so that the year-end
distribution is f(48,100), the process would be
dS = 8dt + 10dz

14
 Itô Process (See pages 306)
In an Itô process the drift rate and the
variance rate are functions of time
dx=a(x,t) dt+b(x,t) dz
The discrete time equivalent
x a ( x, t ) t  b( x, t ) t

is true in the limit as Dt tends to zero

15
Why a Wiener Process Is Not
Appropriate for Stocks
Wiener processes can become negative!
Empirical evidence shows that expected
percentage changes of stock prices in a
short period of time remains constant (not
its expected actual change)
We can reasonably expect that our
uncertainty as to the size of future stock
price movements is proportional to the
level of the stock price 16
An Itô Process for Stock Prices
(See pages 306-309)

dS S dt  S dz

where m is the expected return and s is the

volatility.
The discreteStime
St  S t
equivalent is

The process is known as geometric Brownian

motion
17
Monte Carlo Simulation
We can sample random paths for the
stock price by sampling values for e
Suppose m= 0.15, s= 0.30, and Dt = 1
week (=1/52 or 0.192 years), then
S 0.15 0.0192 S  0.30  0.0192 ε S
or
S 0.00288S  0.0416 S 

18
Example

19
Monte Carlo Simulation – Sampling one
Path (See Table 14.1, page 309)
Stock Price at Random Change in Stock
Week Start of Period Sample for  Price, S

0 100.00 0.52 2.45

1 102.45 1.44 6.43
2 108.88 −0.86 −3.58
3 105.30 1.46 6.70
4 112.00 −0.69 −2.89

20
Itô’s Lemma (See pages 311-313)
If we know the stochastic process followed
by x, Itô’s lemma tells us the stochastic
process followed by some function G (x, t ).
When dx=a(x,t) dt+b(x,t) dz then
G G  2G 2  G
dG  a ½ b  dt  b dz
x t x x
2

Since a derivative is a function of the price of
the underlying asset and time, Itô’s lemma
plays an important part in the analysis of
derivatives

21
Intuition for Itô’s Lemma
A Taylor’s series expansion of G(x, t)
gives
G G  2G
G  x  t  ½ 2 x 2
x t x
 2G  2G 2
 x t  ½ 2 t  
xt t

22
Two applications of Ito lemma
1. The forward price of a stock for a contract
maturing at time T
G S e r (T  t )
dG (  r )G dt  G dz

2. The log of a stock price

G ln S
 2 
dG     dt   dz

 2 

23
The Black-Scholes-Merton
Model

24
 The Stock Price Assumption
Consider a stock whose price is S
In a short period of time of length Dt, the return on
the stock is normally distributed:
S
S
 
 t ,  2 t

where m is expected return and s is the

volatility
The stock price process is assumed to follow a
geometric Brownian motion
25
The Lognormal Property
(Equations 15.2 and 15.3, page 320)

It follows from this assumption that

 2  
ln S T  ln S 0      T ,  T 
2

  2  
or
  2  
ln S T   ln S 0     T ,  T 
2

  2  

Since the logarithm of ST is normal, ST is

lognormally distributed

26
The Lognormal Distribution

E ( ST ) S0 e T
2 2 T 2T
var ( ST ) S0 e (e  1)

27
 The Volatility
The volatility is the standard deviation of the
continuously compounded rate of return in 1
year
The standard deviation of the return in a short
time period time Dt is approximately  t
If a stock price is $50 and its volatility is 25%
per year what is the standard deviation of the
price change in one day?

28
 Estimating Volatility from Historical Data (page 324-326)

1. Take observations S0, S1, . . . , Sn at intervals

of t years (e.g. for weekly data t = 1/52)
2. Calculate the continuously compounded
return in each interval as:
 S 
ui ln i 
 Si  1 

3. Calculate the standard deviation, s , of the ui

´s
s
4. The historical volatility estimate is: ˆ 


29
Nature of Volatility (Business Snapshot
15.2, page 327)
Volatility is usually much greater when the
market is open than when it is closed
Overnight and over weekend price changes
are smaller than intraday price changes
For this reason, time is usually measured in
“trading days” not calendar days when
options are valued
It is assumed that there are 252 trading days
in one year for most assets
30
Black-Scholes-Merton – Main Idea

The option price and the stock price depend on the

same underlying source of uncertainty
We can form a portfolio consisting of the stock and
the option which eliminates this source of uncertainty
The portfolio is instantaneously riskless and must
instantaneously earn the risk-free rate
This leads to the Black-Scholes-Merton differential
equation

31
Assumptions for BSM theory
Market assumptions:
There is no arbitrage opportunity
It is possible to borrow/lend any amount of cash at the risk
free rate
It is possible to buy and sell any amount of stock (short
selling allowed)
There is no transaction cost
Asset dynamics assumptions:
Constant deterministic interest rates
The stock price follows a geometric Brownian motion
The stock does not pay dividends

32
The derivation of the BSM equation
𝑑𝑆=𝜇𝑆 𝑑𝑡+𝜎 𝑆 𝑑𝑧

( )
2
𝜕𝑓 𝜕𝑓 1 𝜕 𝑓 2 2 𝜕𝑓
𝑑𝑓 = 𝜇𝑆+ + 𝜎 𝑆 𝑑𝑡 + 𝜎 𝑆𝑑𝑧
𝜕𝑆 𝜕𝑡 2 𝜕𝑆 2
𝜕𝑆

Set up a portfolio consisting of:

-1 : derivative
+𝜕 𝑓
: shares
𝜕𝑆
This gets rid of the dependence on dz

33
The Derivation of the Black-Scholes-Merton
Differential Equation continued

The value of the portfolio, , is given by

ƒ
  ƒ  S
S
The change in its value in time t is given by
ƒ
  ƒ  S
S

34
The derivation of the BSM
differential equation (continued)
The return on the portfolio must be the risk-free rate. Hence
𝑑 Π =𝑟 Π 𝑑𝑡

− 𝑑𝑓 +
𝜕𝑓
𝜕𝑆 (
𝑑𝑆=𝑟 − 𝑓 +
𝜕𝑓
𝜕𝑆 )
𝑆 𝑑𝑡

Substituting dS and df in this equation we get the

Black-Scholes differential equation
2
𝜕𝑓 𝜕𝑓 1 2 2𝜕 𝑓
+𝑟𝑆 + 𝜎 𝑆 =𝑟𝑓
𝜕𝑆 𝜕𝑆 2 𝜕S
2

35
The Differential Equation
Any security whose price is dependent on the
stock price satisfies the differential equation
The particular security being valued is
determined by the boundary conditions of the
differential equation
In a forward contract the boundary condition is
ƒ = S – K when t =T
The solution to the equation is
ƒ = S – K e–r (T – t )

36
Risk-Neutral Valuation
The variable m does not appear in the
Black-Scholes-Merton differential equation
The equation is independent of all variables
affected by risk preference
The solution to the differential equation is
therefore the same in a risk-free world as it
is in the real world
This leads to the principle of risk-neutral
valuation

37
 Applying Risk-Neutral Valuation

1. Assume that the expected return from the stock

price is the risk-free rate
2. Calculate the expected payoff from the option
3. Discount at the risk-free rate

38
Valuing a Forward Contract with Risk-
Neutral Valuation
Payoff is ST – K
Expected payoff in a risk-neutral world is
S0erT – K
Present value of expected payoff is
e-rT[S0erT – K] = S0 – Ke-rT

39
Proving Black-Scholes-Merton Using
Risk-Neutral Valuation (Appendix to Chapter 15)
∞
𝑐=𝑒 −𝑟𝑇
𝐸 [ max ( 𝑆𝑇 − 𝐾 , 0 ) ]=𝑒 − 𝑟𝑇
∫ max ( 𝑆𝑇 − 𝐾 , 0 ) 𝑔 ( 𝑆 𝑇 ) 𝑑 𝑆𝑇
𝐾

where g(ST) is the probability density function for the lognormal

distribution of ST in a risk-neutral world. Recall that ln(ST /S0) is
𝑚=( 𝑟 −1 /2 𝜎 2 ) 𝑇 , 𝑠=𝜎 √ 𝑇
normally distributed like f(m, s ) with
2

ln ST  m
Q
Change the integration variable to s


 rT
c e max(e Qs m  K , 0)h(Q)dQ
so that (ln K  m ) / s

where h is the probability density function for a standard normal.

Evaluating the integral leads to the BS equation.

40
The Black-Scholes-Merton
Formulas for Options (See pages 333-334)
c S 0 N (d1 )  K e  rT N (d 2 )
p K e  rT N ( d 2 )  S 0 N ( d1 )
ln( S 0 / K )  (r   2 / 2)T
where d1 
 T
ln( S 0 / K )  (r   2 / 2)T
d2  d1   T
 T

41
The N(x) Function
N(x) is the probability that a normally distributed
variable with a mean of zero and a standard deviation
of 1 is less than x

Can be evaluated in Excel with NORMDIST(x,0,1,1)

42
Properties of Black-Scholes Formula

As S0 becomes very large c tends to S0 – Ke-rT

and p tends to zero
As S0 becomes very small c tends to zero and
p tends to Ke-rT – S0
What happens as s becomes very large?
What happens as T becomes very large?

43
Understanding Black-Scholes
c e N (d )S e N d  N d  K 
 rT
2 0
rT
1 2

e  rT : Present value factor

N (d 2 ) : Probability of exercise
S 0 e rT N (d1 )/N (d 2 ) : Expected stock price in a risk - neutral world
if option is exercised
K : Strike price paid if option is exercised

44
Implied Volatility
The implied volatility of an option is the
volatility for which the Black-Scholes-Merton
price equals the market price
There is a one-to-one correspondence
between prices and implied volatilities
Traders and brokers often quote implied
volatilities rather than dollar prices

45
The VIX S&P500 Volatility Index

46
Dividends
European options on dividend-paying stocks
are valued by substituting the stock price less
the present value of dividends into Black-
Scholes
Only dividends with ex-dividend dates during
life of option should be included
The “dividend” should be the expected
reduction in the stock price expected

47