8: The Black-Scholes Model

Marek Rutkowski
School of Mathematics and Statistics
University of Sydney

MATH3075/3975 Financial Mathematics

Semester 2, 2016

8: The Black-Scholes Model

Outline

We will examine the following issues:

1 The Wiener Process and its Properties
2 The Black-Scholes Market Model
3 The Black-Scholes Call Option Pricing Formula
4 The Black-Scholes Partial Differential Equation
5 Random Walk Approximations

8: The Black-Scholes Model

 PART 1

THE WIENER PROCESS AND ITS PROPERTIES

8: The Black-Scholes Model

The Origin of the Wiener Process

The Brownian motion is a mathematical model used to

describe the random mouvements of particles. It was named
after Scottish botanist Robert Brown (1773-1858) who has
published in 1827 a paper in which the chaotic mouvements
of pollen suspended in water were examined.
The Brownian motion was used by Louis Bachelier in his PhD
thesis completed in 1900 and devoted to pricing of options.
The Brownian motion was also used by physicists to describe
the diffusion mouvements of particles, in particular, by Albert
Einstein (1879-1955) in his famous paper published in 1905.
The Brownian motion is also known as the Wiener process
in honour of the famous American mathematician Norbert
Wiener (1894-1964).
The Brownian motion is nowadays widely used to model
uncertainty in engineering, economics and finance.
8: The Black-Scholes Model
Wiener Process: Definition

Definition (Wiener Process)

A stochastic process W = (Wt , t ∈ R+ ) is called the Wiener
process (or the standard Brownian motion) if the following
conditions hold:
1 W0 = 0.
2 Sample paths of the process W , that is, the maps t → Wt (ω)
are continuous functions.
3 The process W has the Gaussian (i.e. normal) distribution
with the expected value EP (Wt ) = 0 for all t ≥ 0 and the
covariance

Cov (Ws , Wt ) = min (s, t), s, t ≥ 0.

8: The Black-Scholes Model

Wiener Process: Equivalent Definition

Definition (Wiener Process: Equivalent Definition)

A stochastic process W = (Wt , t ∈ R+ ) on Ω is called the
Wiener process if the following conditions hold:
1 W0 = 0.
2 Sample paths of W are continuous functions.
3 For any 0 ≤ s < t, Wt − Ws is normally distributed with
mean 0 and variance t − s.
4 For any 0 ≤ t1 < t2 < · · · < tn ,

Wt1 , Wt2 − Wt1 , . . . , Wtn − Wtn−1

are mutually independent.

8: The Black-Scholes Model

Existence of the Wiener Process

The existence of a stochastic process satisfying the definition

of a Wiener process is not obvious.
The following theorem was first rigorously established by
Norbert Wiener in his paper published in 1923.

Theorem (Wiener’s Theorem)

There exists a probability space (Ω, F, P) and a process W defined
on this space, such that conditions 1)-3) of the definition of the
Wiener process are met.

It is known that almost all sample paths of the Wiener process

are continuous functions of the time parameter, but they are
non-differentiable everywhere. This striking feature makes the
Wiener process rather difficult to analyse.

8: The Black-Scholes Model

Wiener Process: Sample Paths

Simulation of a Wiener process

2

−1

−2

−3

−4
0 1 2 3 4 5
Time

Figure: Three sample paths of a Wiener process with ∆t = 0.005

8: The Black-Scholes Model

Gaussian Distribution

Remark (Gaussian Distribution)

We say that X has the Gaussian (normal) distribution with
mean µ ∈ R and variance σ 2 > 0 if its pdf equals

1 (x−µ)2
f (x) = √ e − 2σ2 for x ∈ R.
2πσ 2

We write X ∼ N(µ, σ 2 ).
One can show that
Z ∞ Z ∞
1 (x−µ)2
f (x) dx = √ e− 2σ 2 dx = 1.
−∞ −∞ 2πσ 2
We have
EP (X ) = µ and Var (X ) = σ 2 .

8: The Black-Scholes Model

Standard Normal Distribution

Remark (Standard Normal Distribution)

If we set µ = 0 and σ 2 = 1 then we obtain the standard
normal distribution N(0, 1) with the following pdf

1 x2
n(x) = √ e − 2 for x ∈ R.
2π
The cdf of the probability distribution N(0, 1) equals
Z x Z x
1 u2
N(x) = n(u) du = √ e − 2 du for x ∈ R.
−∞ −∞ 2π

The values of N(x) can be found in the cumulative standard

normal table (also known as the Z table).


If X ∼ N µ, σ 2 then Z := X σ−µ ∼ N(0, 1).

8: The Black-Scholes Model

Marginal Distributions of the Wiener Process

Let N(µ, σ 2 ) denote the Gaussian (normal) distribution with

mean µ and variance σ 2 .
√
For any t > 0, Wt ∼ N(0, t) and thus ( t)−1 Wt ∼ N(0, 1).
The random variable Wt has the pdf p(x, t) given by

1 2
p(t, x) = √ e −x /2t , for x ∈ R.
2πt
Hence for any real numbers a ≤ b
Z b Z √b
1 2 t 1 2
P(Wt ∈ [a, b]) = √ e −x /2t dx = √ e −x /2 dx
a 2πt √a
t
2π
Z b    
√
t b a
= n(x) dx = N √ − N √ .
a
√
t
t t

8: The Black-Scholes Model

Markov Property (MATH3975)
Proposition (8.1)
The Wiener process W is a Markov process in the following sense:
for every n ≥ 1, any sequence of times 0 < t1 < . . . < tn < t and
any real numbers x1 , . . . , xn , the following holds for all x ∈ R

P ( Wt ≤ x| Wt1 = x1 , . . . , Wtn = xn ) = P ( Wt ≤ x| Wtn = xn ) .

Moreover, for all s < t and x, y ∈ R we have

Z y
P ( Wt ≤ y | Ws = x) = p(t − s, z − x) dz
−∞

where
 
1 (z − x)2
p(t − s, z − x) = p exp −
2π(t − s) 2(t − s)

is the transition probability density function of the Wiener process.

8: The Black-Scholes Model
Martingale Property (MATH3975)

Proposition (8.2)
Let W be the Wiener process on a probability space (Ω, F, P).
Then the process W is a martingale with respect to its natural
filtration Ft = FtW , that is, the filtration generated by W .

Proof of Proposition 8.2.

For all 0 ≤ s < t, using the independence of increments of the
Wiener process W , we obtain


EP (Wt | Fs ) = EP (Wt − Ws ) + Ws | Fs
= EP (Wt − Ws ) + Ws
= Ws .

We conclude that W is a martingale with respect to its natural

filtration.

8: The Black-Scholes Model

 PART 2

THE BLACK-SCHOLES MARKET MODEL

8: The Black-Scholes Model

Stock Price Process

We note that the values of the Wiener process W can be

negative and thus it cannot be used to directly model the
movements of the stock price.
Following Samuelson (1965) and Black and Scholes (1973),
we postulate that the stock price process S is governed under
e by the following
the risk-neutral probability measure P
stochastic differential equation (SDE)

dSt = r St dt + σSt dWt (1)

with a constant initial value S0 > 0.

The term σSt dWt is aimed to give a plausible description of
the uncertainty of the stock price.
The volatility parameter σ > 0 is used to control the size of
random fluctuations of the stock price.

8: The Black-Scholes Model

Stochastic Differential Equation

Sample path of the Wiener process W are not differentiable

so that equation (1) cannot be represented as

dSt = r St dt + σSt Wt′ dt.

It should be understood as the stochastic integral equation

Z t Z t
St = S0 + rSu du + σSu dWu
0 0

where the second integral is the Itō stochastic integral.

The Itō stochastic integration theory, which extends the
classic integrals and underpins financial modelling in
continuous time, is beyond the scope of this course.

8: The Black-Scholes Model

The Black-Scholes Model

It turns out that stochastic differential equation (1) can be

solved explicitly yielding the unique solution
  
1 2
St = S0 exp σWt + r − σ t . (2)
2

The process S is called the geometric Brownian motion.

Note that St has the lognormal distribution for every t > 0.
It can be shown that S is a Markov process. Note, however,
that S is not a process of independent increments.
We assume that the continuously compounded interest rate r
is constant. Hence the savings account equals

Bt = B0 e rt , t ≥ 0,

where B0 = 1. Hence dBt = rBt dt for t ≥ 0.

8: The Black-Scholes Model

Sample Paths of Stock Price

75

70

Simulation of a solution to SDE (2) 65

60

55

50

45

40

35
0 0.2 0.4 0.6 0.8 1
Time

Figure: Three sample paths of the stock price with r = 10%, σ = 0.2
and ∆t = 0.001

8: The Black-Scholes Model

The Black-Scholes Model M = (B, S)

Assumptions of the Black-Scholes market model M = (B, S):

There are no arbitrage opportunities in the class of trading
strategies.
It is possible to borrow or lend any amount of cash at a
constant interest rate r ≥ 0.
The stock price dynamics are governed by a geometric
Brownian motion.
It is possible to purchase any amount of a stock and
short-selling is allowed.
The market is frictionless: there are no transaction costs (or
any other costs).
The underlying stock does not pay any dividends.

8: The Black-Scholes Model

Discounted Stock Price (MATH3975)

As in a multi-period market model, the discounted stock price Ŝ is

a martingale.
Proposition (8.3)
The discounted stock price, that is, the process Ŝ given by the
formula
St
Ŝt = = e −rt St
Bt
e that
is a martingale with respect to its natural filtration under P,
is, for every 0 ≤ s ≤ t,


EPe Ŝt  Ŝu , u ≤ s = Ŝs .

8: The Black-Scholes Model

Proof of Proposition 8.3 (MATH3975)

Proof of Proposition 8.3.

We observe that equality (2) yields
1 2t 1 2 (t−s)
Ŝt = S0 e σWt − 2 σ = Ŝs e σ(Wt −Ws )− 2 σ . (3)

Hence if we know Ŝt then we also know the value of Wt and

vice versa. This immediately implies that FŜ = FW .
Therefore, the following conditional expectations coincide



EPe X  Ŝu , u ≤ s = EPe X | Wu , u ≤ s (4)

for any integrable random variable X

8: The Black-Scholes Model

Proof of Proposition 8.3 (MATH3975)

Proof of Proposition 8.3 (Continued).

We obtain the following chain of equalities


EeP Ŝt  Ŝu , u ≤ s
  
= EeP Ŝs e σ(Wt −Ws − 2 σ (t−s))  Ŝu , u ≤ s
1 2
(from (3))
1 2
  
= Ŝs e − 2 σ (t−s) EeP e σ(Wt −Ws )  Ŝu , u ≤ s (conditioning)
1 2
  
= Ŝs e − 2 σ (t−s) EeP e σ(Wt −Ws )  Wu , u ≤ s (from (4))
1 2
 
= Ŝs e − 2 σ (t−s) EeP e σ(Wt −Ws ) . (independence)

It remains to compute the expected value above.

8: The Black-Scholes Model

Proof of Proposition 8.3 (MATH3975)

Proof of Proposition 8.3 (Continued).

√
Recall also that Wt − Ws = t − s Z where Z ∼ N(0, 1), and
thus

1 2 √

EPe Ŝt  Ŝu , u ≤ s = Ŝs e − 2 σ (t−s) EPe e σ t−sZ .

Let us finally observe that if Z ∼ N(0, 1) then for any real a

2
EPe e aZ = e a /2 .
√
By setting a = σ t − s, we finally obtain

1 2 1 2
EPe Ŝt  Ŝu , u ≤ s = Ŝs e − 2 σ (t−s) e 2 σ (t−s) = Ŝs

which shows that Ŝ is indeed a martingale.

8: The Black-Scholes Model

 PART 3

THE BLACK-SCHOLES CALL OPTION

PRICING FORMULA

8: The Black-Scholes Model

Call Option

Recall that the European call option written on the stock

is a traded security, which pays at its maturity T the random
amount
CT = (ST − K )+
where x + = max (x, 0) and K > 0 is a fixed strike.
We take for granted that for t ≤ T the price Ct (x) of the call
option when St = x is given by the risk-neutral pricing
formula


Ct (x) = e −r (T −t) EPe (ST − K )+  St = x .

This formula can be supported by the replication principle.

However, this argument requires the knowledge of the Itō
stochastic integration theory with respect to the Brownian
motion, as was developed by Kiyoshi Itō (1944).

8: The Black-Scholes Model

The Black-Scholes Call Pricing Formula

The following call option pricing result was established in the

seminal paper by Black and Scholes (1973).

Theorem (8.1)
The arbitrage price of the call option at time t ≤ T equals



Ct (St ) = St N d+ (St , T − t) − Ke −r (T −t) N d− (St , T − t)

where

ln SKt + r ± 12 σ 2 (T − t)
d± (St , T − t) = √
σ T −t
and N is the standard normal cumulative distribution function.

8: The Black-Scholes Model

Proof of Theorem 8.1 (MATH3975)

Proof of Theorem 8.1.

Our goal is to compute the conditional expectation


Ct (x) = e −r (T −t) EPe (ST − K )+  St = x .

We can represent the stock price ST as follows

1 2
ST = St e (r − 2 σ )(T −t)+σ(WT −Wt ) .

As in the proof of Proposition 8.3, we write

√
WT − Wt = T − tZ

where Z has the standard Gaussian probability distribution,

that is, Z ∼ N(0, 1).

8: The Black-Scholes Model

Proof of Theorem 8.1 (MATH3975)

Proof of Theorem 8.1 (Continued).

Using the independence of increments of the Wiener process
W , we obtain, for a generic value x > 0 of the stock price St
at time t
 +  

St e (r − 2 σ )(T −t)+σ(WT −Wt ) − K
1 2
Ct (x) = e −r (T −t) EeP  St = x
 √ +
= e −r (T −t) EeP x e (r − 2 σ )(T −t)+σ T −tZ − K
1 2

Z ∞ √ +
x e (r − 2 σ )(T −t)+σ T −tz − K
1 2
= e −r (T −t) n(z) dz
−∞

We denote here by n the pdf of Z , that is, the standard

normal probability density function.

8: The Black-Scholes Model

Proof of Theorem 8.1 (MATH3975)

Proof of Theorem 8.1 (Continued).

It is clear that the function under the integral sign is non-zero
if and only if the following inequality holds
1 2
√
x e (r − 2 σ )(T −t)+σ T −tz
− K ≥ 0.

This in turn is equivalent to the following inequality

ln Kx − r − 21 σ 2 (T − t)
z≥ √ = −d− (x, T − t).
σ T −t
Let us denote

d+ = d+ (x, T − t), d− = d− (x, T − t).

8: The Black-Scholes Model

Proof of Theorem 8.1 (MATH3975)
Proof of Theorem 8.1 (Continued).
Z ∞  √ 
x e (r − 2 σ )(T −t)+σ
1 2
−r (T −t) T −tz
Ct (x) = e − K n(z) dz
−d−
Z ∞ √
Z ∞
− 21 σ2 (T −t) σ T −tz −r (T −t)
=xe e n(z) dz − Ke n(z) dz
−d− −d−
Z ∞
1 2
√ 1 1 2
= x e− 2 σ (T −t)
eσ T −tz
√ e − 2 z dz − Ke −r (T −t) N (d− )
−d− 2π
Z ∞
1 √ 2
√ e − 2 (z−σ T −t ) dz − Ke −r (T −t) N (d− )
1
=x
−d− 2π
Z ∞
1 2
=x √
√ e −u /2 du − Ke −r (T −t) N (d− )
−d− −σ T −t 2π
Z ∞
1 2
=x √ e −u /2 du − Ke −r (T −t) N (d− )
−d+ 2π
= xN (d+ ) − Ke −r (T −t) N (d− ) .

8: The Black-Scholes Model

Put-Call Parity

The price of the put option can be computed from the usual
put-call parity

Ct − Pt = St − Ke −r (T −t) = St − KB(t, T ).

Specifically, the put option price equals

Pt = Ke −r (T −t) N − d− (St , T − t) − St N − d+ (St , T − t) .

It is worth noting that Ct > 0 and Pt > 0.

It can also be checked that the prices of call and put options
are increasing functions of the volatility parameter σ (if all
other quantities are fixed). Hence the options become more
expensive when the underlying stock becomes more risky.
The price of a call (put) option is an increasing (decreasing)
function of the interest rate r .

8: The Black-Scholes Model

Example: Call Option

Example (8.1)
Suppose that the current stock price equals $31, the stock
price volatility is σ = 10% per annum, and the risk-free rate
is r = 5% per annum with continuous compounding.
Consider a call option on the stock S, with strike price $30
and with 3 months to expiry. We may assume that t = 0
and T = 0.25. We obtain d+ (S0 , T ) = 0.93 and thus
√
d− (S0 , T ) = d+ (S0 , T ) − σ T = 0.88.

The Black-Scholes call option pricing formula yields

(approximately)

C0 = 31N(0.93) − 30e −0.05/4 N(0.88) = 25.42 − 23.9 = 1.52

since N(0.93) ≈ 0.82 and N(0.88) ≈ 0.81.

8: The Black-Scholes Model

Replicating Strategy

Example (8.1 Continued)

Let Ct = φ0t Bt + φ1t St . The hedge ratio for the call option is
known to be given by the formula

φ1t = N(d+ (St , T − t)).

Hence the replicating portfolio at time t = 0 is given by

φ00 = −23.9, φ10 = N(d+ (S0 , T )) = 0.82.

This means that to hedge a short position in the call option,

which was sold at the arbitrage price C0 = $1.52, the writer
needs to buy at time 0 the number δ = 0.82 shares of stock.
The purchase of shares requires an additional borrowing of
23.9 units of cash.

8: The Black-Scholes Model

Elasticity of the Call Price

Example (8.1 Continued)

The elasticity at time 0 of the call option price with respect
to the stock price equals
 

c ∂C C0 −1 N d+ (S0 , T ) S0
η0 := = = 16.72.
∂S S0 C0

Suppose that the stock price rises immediately from $31 to

$31.2, yielding a return rate of 0.65% flat.
Then the option price will move by approximately 16.5 cents
from $1.52 to $1.685, giving a return rate of 10.86% flat.
The option has nearly 17 times the return rate of the stock;
this also means that it will drop 17 times as fast.

8: The Black-Scholes Model

Example: Put Option
Example (8.1 Continued)
We now assume that an option is a put. The price of the put
option at time 0 equals

P0 = 30e −0.05/4 N(−0.88) − 31N(−0.93) = 5.73 − 5.58 = 0.15

The hedge ratio corresponding to a short position in the put

option equals approximately −0.18 (since N(−0.93) ≈ 0.18).
Therefore, to hedge the exposure an investor needs to short
0.18 shares of stock for one put option. The proceeds from
the option and share-selling transactions, which amount to
$5.73, should be invested in risk-free bonds.
Notice that the elasticity of the put option is several times
larger than the elasticity of the call option. For instance,
if the stock price rises immediately from $31 to $31.2 then
the price of the put option will drop to less than 12 cents.
8: The Black-Scholes Model
 PART 4

THE BLACK-SCHOLES PDE

8: The Black-Scholes Model

The Black-Scholes PDE

Proposition (8.4)
Consider a path-independent contingent claim X = g (ST ). Let
the price of the contingent claim at t given the current stock price
St = s be denoted by v (s, t). Then v (s, t) is the solution of
the Black-Scholes partial differential equation

∂ σ2 s 2 ∂ 2 ∂
v (s, t) + 2
v (s, t) + rs v (s, t) − rv (s, t) = 0
∂t 2 ∂s ∂s
with the terminal condition v (s, T ) = g (s).

Proof of Proposition 8.4.

The statement is an immediate consequence of the risk-neutral
valuation formula and the classic Feynman-Kac formula.

8: The Black-Scholes Model

Sensitivities of the Call Price
It can be checked that arbitrage prices of call and put options
satisfy this PDE.
We denote by c(s, τ ) the function c : R+ × [0, T ] → R such
that Ct = c(St , T − t). Then

cs = N(d+ ) = δ > 0,
n(d+ )
css = √ = γ > 0,
sσ τ
sσ
cτ = √ n(d+ ) + Kre −r τ N(d− ) = θ > 0,
2 τ
√
cσ = s τ n(d+ ) = λ > 0,
cr = τ Ke −r τ N(d− ) = ρ > 0,
cK = −e −r τ N(d− ) < 0,

where d+ = d+ (s, τ ), d− = d− (s, τ ) and n stands for the

standard Gaussian probability density function.
8: The Black-Scholes Model
Sensitivities of the Put Price

We denote by p(s, τ ) the function p : R+ × [0, T ] → R such

that Pt = p(St , T − t). Then

ps = N(d+ ) − 1 = −N(−d+ ) = δ < 0,

n(d+ )
pss = √ = γ > 0,
sσ τ
sσ
pτ = √ n(d+ ) + Kre −r τ (N(d− ) − 1) = θ,
2 τ
√
pσ = s τ n(d+ ) = λ > 0,
pr = τ Ke −r τ (N(d− ) − 1) = ρ < 0,
pK = e −r τ (1 − N(d− )) > 0.

where d+ = d+ (s, τ ), d− = d− (s, τ ) and n stands for the

standard Gaussian probability density function.

8: The Black-Scholes Model

 PART 5

RANDOM WALK APPROXIMATIONS

8: The Black-Scholes Model

Random Walk Approximations

Our final goal is to examine a judicious approximation of the

Black-Scholes model by a sequence of CRR models.
In the first step, we will first examine an approximation of the
Wiener process by a sequence of symmetric random walks.
In the next step, we will use this result in order to show how
to approximate the Black-Scholes stock price process by a
sequence of the CRR stock price models.
We will also recognise that the proposed approximation of
the stock price leads to the Jarrow-Rudd parametrisation of
the CRR model in terms of the short term rate r and the
stock price volatility σ.

8: The Black-Scholes Model

Symmetric Random Walk

Definition (Symmetric Random Walk)

A process Y = (Yk , k = 0, 1, . . . ) on a probability space (Ω, F, P)
is called the
Psymmetric random walk starting at zero if Y0 = 0
k
and Yk = i =1 Xi where the random variables X1 , X2 , . . . are
independent with the following common probability distribution

P(Xi = 1) = 0.5 = P(Xi = −1).

The scaled h
√ random walk Y is obtained from Y as follows: we
fix h = ∆t and for every k = 0, 1, . . . we set

√ k √
X
h
Yk∆t = ∆tYk = ∆tXi
i =1
√
Of course, for h = h
∆t = 1 we obtain Yk∆t = Yk1 = Yk .
8: The Black-Scholes Model
Scaled Random Walk

1.5

Y(3∆ t) = 3h
1
0.5
Y(2∆ t) = 2h
0.5
0.5 Y(∆ t) = h
0.5 ...
0.5 0.5 0.5
Y(3∆ t) = h
0
...
0.5 0.5
Y(0) = 0 Y(2∆ t) = 0
0.5
−0.5 Y(∆ t) = −h
0.5
Y(3∆ t) = −h
...
0.5
Y(2∆ t) = −2h
0.5
−1
Y(3∆ t) = −3h

−1.5

Figure: Representation of the scaled random walk Y h

8: The Black-Scholes Model

Approximation of the Wiener Process

The following result is an easy consequence of the classic

Central Limit Theorem (CLT) for sequences of independent
and identically distributed (i.i.d.) random variables.

Theorem (8.2)
Let Yth for t = 0, ∆t,
√ . . . , be a random walk starting at 0 with
increments ±h = ± ∆t. If



h
P Yt+∆t = y + h Yth = y = P Yt+∆t
h
= y − h Yth = y = 0.5

then, for any fixed t ≥ 0, the limit limh→0 Yth exists in the sense
of probability distribution. Specifically, limh→0 Yth ∼ Wt where W
is the Wiener process and ∼ denotes the equality of probability
distributions. In other words, limh→0 Yth ∼ N(0, t).

8: The Black-Scholes Model

Proof of Theorem 8.2
Proof of Theorem 8.2.
We fix t > 0 and we set k =√t/∆t. Hence if ∆t → 0 then
k → ∞. We recall that h = ∆t and
k √
X
h
Yk∆t = ∆tXi .
i =1

Since EP (Xi ) = 0 and Var (Xi ) = EP (Xi2 ) = 1, we obtain

√ k
X
h
EP (Yk∆t )= ∆t EP (Xi ) = 0
i =1

k
X k
X
h
Var (Yk∆t )= ∆t Var (Xi ) = ∆t = k∆t = t.
i =1 i =1

Hence the statement follows from the (slightly extended) CLT.

8: The Black-Scholes Model

Central Limit Theorem (CLT)

Theorem (Central Limit Theorem)

Assume that X1 , X2 , . . . are independent and identically distributed
random variables with mean µ and variance σ 2 > 0. Then for all
real x
  Z x
X1 + · · · + Xn − nµ 1 2
lim P √ ≤x = √ e −u /2 du = N(x).
n→∞ σ n −∞ 2π

Pn
Note that if we denote Yn = i =1 Xi then

X1 + · · · + Xn − nµ Yn − EP (Yn )
√ = p .
σ n Var (Yn )

8: The Black-Scholes Model

Approximation of the Wiener Process

The sequence of random h

√ walks Y approximates the Wiener
process W when h = ∆t → 0 meaning that:
For any fixed t ≥ 0, the convergence limh→0 Yth ∼ Wt holds,
where ∼ denotes the equality of probability distributions on R.
This follows from Theorem 8.2.
For any fixed n and any dates 0 ≤ t1 < t2 < · · · < tn , we have

lim (Yth1 , . . . , Ythn ) ∼ (Wt1 , . . . , Wtn )

h→0

where ∼ denotes the equality of probability distributions on

the space Rn . This can be proven by extending Theorem 8.2.
The sequence of linear versions of the random walk processes
Y h converge to a continuous time process W in the sense of
the weak convergence of stochastic processes on the space of
continuous functions (this is due to Donsker (1951)).

8: The Black-Scholes Model

Approximation of the Stock Price

Recall that the JR parameterisation for the CRR binomial

model postulates that
 2
 √  2
 √
r − σ2 ∆t + σ ∆t r − σ2 ∆t − σ ∆t
u=e and d = e ,
√
whereas under the CRR convention we set u = e σ ∆t = 1/d.
We will show that it corresponds to a particular approximation
of the stock price process S
Recall that for all 0 ≤ s ≤ t
 2

r − σ2 t+σWt
St = S0 e
 2
  2

r − σ2 s+σWs r − σ2 (t−s)+σ(Wt −Ws ) (5)
= S0 e e
 2

r − σ2 (t−s)+σ(Wt −Ws )
= Ss e .

8: The Black-Scholes Model

Approximation of the Stock Price
Let us set t − s = ∆t and let us replace the Wiener process
W by the random walk Y h in equation (5). Then
h
√
Wt+∆t − Wt ≈ Yt+∆t − Yth = ±h = ± ∆t.

Consequently, we obtain the following approximation

  2
 √
 r − σ2 ∆t + σ ∆t
St e  if the price increases,
St+∆t ≈ 2
 √
 r − σ2 ∆t − σ ∆t
St e if the price decreases.
More explicitly, for k = 0, 1, . . .
 2

r − σ2 h
k∆t + σYk∆t
h
Sk∆t = S0 e .

If we denote t = k∆t then

 2

r − σ2 t+σYth
Sth = S0 e .

8: The Black-Scholes Model

Jarrow-Rudd Parametrisation

We observe that this approximation of the stock price process

leads to the Jarrow-Rudd parameterisation
 2
 √  2
 √
r − σ2 ∆t + σ ∆t r − σ2 ∆t − σ ∆t
u=e and d = e .

The convergence of the sequence of random walks Y h to

the Wiener process W (Donsker’s Theorem) implies that
the sequence S h of CRR stock price models converges to
the Black-Scholes stock price model S.
The convergence of S h to the stock price process S justifies
the claim that the JR parametrisation is more suitable than
the CRR method.
This is especially important when dealing with valuation and
hedging of path-dependent and American contingent claims.

8: The Black-Scholes Model

THE END

8: The Black-Scholes Model