20912lectures01 19

Lectures
Sergei Fedotov
20912 - Introduction to Financial Mathematics
No tutorials in the first week
Sergei Fedotov (University of Manchester) 20912 2010 1/1

Lecture 1
1 Introduction
Elementary economics background
What is financial mathematics?
The role of SDEs and PDEs
2 Time Value of Money
3 Continuous Model for Stock Price

General Information
Textbooks:
J. Hull, Options, Futures and Other Derivatives, 7th Edition,

Prentice-Hall, 2008.
P. Wilmott, S. Howison and J. Dewynne, The Mathematics of Financial
Derivatives: A Student Introduction, Cambridge University Press, 1995
Assessment:
Test in a week 6: 20%

2 hours examination: 80%

Elementary Economics Background
This course is concerned with mathematical models for financial markets:
Stock Markets, such as NYSE(New York Stock Exchange), London

Stock Exchange, etc.
Bond Markets, where participants buy and sell debt securities.
Futures and Option Markets, where the derivative products are traded.
Example: European call option gives the holder the right (not obligation)
to buy underlying asset at a prescribed time T for a specified price E .
Option market is massive! More money is invested in options than in the

underlying securities. The main purpose of this course is to determine the
price of options.
Why stochastic differential equations (SDEs) and partial differential

equations (PDEs)?
Time value of money
What is the future value V (t) at time t = T of an amount P invested or
borrowed today at t = 0?
Simple interest rate:
V (T ) = (1 + rT )P (1)
where r > 0 is the simple interest rate, T is expressed in years.
Compound interest rate:
r mT
V (T ) = 1 + P (2)
m
where m is the number interest payments made per annum.
Continuous compounding:
In the limit m , we obtain
V (T ) = e rT P (3)
1 z

since e = limz 1 + z . Throughout this course the interest rate r
will be continuously compounded.
Simple Model for Stock Price S(t)
Let S(t) represent the stock price at time t. How to write an equation for
this function?
Return (relative measure of change):
S
(4)
S
where S = S(t + t) S(t)
In the limit t 0 :
dS
(5)
S
How to model the return?

Let us decompose the return into two parts: deterministic and stochastic

Modelling of Return
Return:
dS
= dt + dW (6)
S
where is a measure of the expected rate of growth of the stock price. In
general, = (S, t). In simpe models is taken to be constant
( = 0.1 0.3).
dW describes the stochastic change in the stock price, where dW

stands for
W = W (t + t) W (t)
as t 0
W (t) is a Wiener process
is the volatility ( = 0.2 0.5)

Stochastic differential equation for stock price
dS = Sdt + SdW (7)
Simple case: volatility = 0
dS = Sdt (8)

Wiener process
Definition. The standard Wiener process W (t) is a Gaussian process such
that
W (t) has independent increments: if u v s t, then W (t) W (s)

and W (v ) W (u) are independent
W (s + t) W (s) is N(0, t) and W (0) = 0
Clearly
EW (t) = 0 and EW 2 = t, where E is the expectation operator.
The increment W = W (t + t) W (t) can be written as

1
W = X (t) 2 , where X is a random variable with normal distribution
with zero mean and unit variance:
X N (0, 1)
EW = 0 and E(W )2 = t.
Lecture 2
1 Properties of Wiener Process
2 Approximation for Stock Price Equation
3 Itos Lemma
Sergei Fedotov (University of Manchester) 20912 2010 10 / 1

Wiener process
The probability density function for W (t) is
2
1 y
p(y , t) = exp
2t 2t
Rb
and P (a W (t) b) = a p(y , t)dy
Simulations of a Wiener process:

Approximation of SDE for small t
The increment W = W (t + t) W (t) can be written as

1
W = X (t) 2 , where X is a random variable with normal distribution
with zero mean and unit variance: X N (0, 1)
EW = 0 and E(W )2 = t.
Recall: equation for the stock price is
dS = Sdt + SdW ,
then 1
S St + SX (t) 2
It means S N St, 2 S 2 t


Examples
Example 1. Consider a stock that has volatility 30% and provides expected
return of 15% p.a. Find the increase in stock price for one week if the
initial stock price is 100.
Answer: S = 0.288 + 4.155X
Note: 4.155 is the standard deviation

S
Example 2. Show that the return S is normally distributed with mean
t and variance 2 t

Itos Lemma
We assume that f (S, t) is a smooth function of S and t.
Find df if dS = Sdt + SdW
Volatility = 0
f f f f

df = t dt + S dS = t + S S dt
Volatility 6= 0
Itos Lemma:

1 2 2 2f
df = f t + S f
S + 2 S S 2
f
dt + S S dW
Example. Find the SDE satisfied by f = S 2 .

Lecture 3
1 Distribution for ln S(t)
2 Solution to Stochastic Differential Equation for Stock Price
3 Examples

Differential of ln S
Example 1. Find the stochastic differential equation (SDE) for
f = ln S
by using Itos Lemma:

1 2 2 2f
df = f t + S f
S + 2 S S 2
f
dt + S S dW
We obtain
2

df = dt + dW
2
This is a constant coefficient SDE.
Integration from 0 to t gives
2

f f0 = t + W (t) since W (0) = 0.
2
Normal distribution for ln S(t)
We obtain for ln S(t)

2

ln S(t) ln S0 = t + W (t)
2
where S0 = S(0) is the initial stock price.

2
ln S(t) has a normal distribution with mean ln S0 + 2 t and
variance 2 t.
Example 2. Consider a stock with an initial price of 40, an expected return

of 16% and a volatility of 20%.
Find the probability distribution of ln S in six months.
We have
2

2
ln S(T ) N ln S0 + T, T
2
Answer: ln S(0.5) N (3.759, 0.020)

Probability density function for ln S(t)
Recall that if the random variable X has a normal distribution with mean
and variance 2 , then the probability density function is
(x )2

1
p(x) = exp
2 2 2 2
The probability density function of X = ln S(t) is

(x ln S0 ( 2 /2)t)2

1
exp
2 2 t 2 2 t
Exact expression for stock price S(t)
Definition. The model of a stock dS = Sdt + SdW is known as a

geometric Brownian motion.
The random function S(t) can be found from
2

ln(S(t)/S0 ) = t + W (t)
2
2

2 t+W (t)
Stock price at time t: S(t) = S0 e
Or
2

2 t+ tX
S(t) = S0 e where X N (0, 1)

Lecture 4
1 Financial Derivatives
2 European Call and Put Options
3 Payoff Diagrams, Short Selling and Profit

Derivatives
Derivative is a financial instrument which value depends on the values of
other underlying variables. Other names are financial derivative, derivative
security, derivative product. A stock option, for example, is a derivative
whose value is dependent on the stock price. Examples: forward contracts,
futures, options, swaps, CDS, etc.
Options are very attractive to investors, both for speculation and for
hedging
What is an Option?
Definition.
European call option gives the holder the right (not obligation) to buy
underlying asset at a prescribed time T for a specified (strike) price E .
European put option gives its holder the right (not obligation) to sell
underlying asset at a prescribed time T for a specified (strike) price E .
The question is what does this actually mean?

Example.
Consider a three-month European call option on a BP share with a strike

price E = 15 (T = 0.25). If you enter into this contract you have the right
but not the obligation to buy one share for E = 15 in a three months time.
Whether you exercise your right depends on the stock price in the market
at time T :
If the stock price is above 15, say 25, you can buy the share for 15,
and sell it immediately for 25, making a profit of 10.
If the stock price is below 15, there is no financial sense to buy it. The
option is worthless.

Payoff Diagrams
We denote by C (S, t) the value of European call option and P(S, t) the
value of European put option
Definition. Payoff Diagram ia a graph of the value of the option position

at expiration t = T as a function of the underlying stock price S.

0, S E ,
Call price at t = T : C (S, T ) = max (S E , 0) =
S E, S > E,

E S, S E ,
Put price at t = T : P(S, T ) = max (E S, 0) =
0, S > E ,
If a trader thinks that the stock price is on the rise, he can make money by
purchase a call option without buying the stock. If a trader believes the
stock price is on the decline, he can make money by buying put options.

Profit of call option buyer
The profit (gain) of a call option holder (buyer) at time T is
max (S E , 0) C0 e rT , C0 initial call price
Example:
Find the stock price on the exercise date in three months, for a European
call option with strike price 10 to give a gain (profit) of 14 if the option
is bought for 2.25, financed by a loan with continuously compounded
interest rate of 5%
Solution:
1
14 = S(T ) 10 2.25 e 0.05 4 ,
S(T ) = 26.28
For the holder of European put option, the profit at time T is
max (E S, 0) P0 e rT
.
Portfolio and Short Selling
Definition. Short selling is the practice of selling assets that have been
borrowed from a broker with the intention of buying the same assets back
at a later date to return to the broker.
This technique is used by investors who try to profit from the falling price
of a stock.
Definition. Portfolio is the combination of assets, options and bonds.
We denote by the value of portfolio. Example: = 2S + 4C 5P.
It means that portfolio consists of long position in two shares, long

position in four call options and short position in five put options.

Option positions

Straddle
Straddle is the purchase of a call and a put on the same underlying
security with the same maturity time T and strike price E .
The value of portfolio is = C + P
Straddle is effective when an investor is confident that a stock price will

change dramatically, but is uncertain of the direction of price move.
Example of large profits: S0 = 40, E = 40, C0 = 2, P0 = 2
Let us find an expected return!!

Ans: 400
Lecture 5
1 Trading Strategies: Straddle, Bull Spread, etc.
2 Bond and Risk-Free Interest Rate
3 No Arbitrage Principle

Portfolio and Short Selling
Reminder from previous lecture 4.
Definition. Short selling is the practice of selling assets that have been
borrowed from a broker with the intention of buying the same assets back
at a later date to return to the broker. This technique is used by investors
who try to profit from the falling price of a stock.
Definition. Portfolio is the combination of assets, options and bonds. We

denote by the value of portfolio.
Examples.

Option positions

Trading Strategies Involving Options
Straddle is the purchase of a call and a put on the same underlying

security with the same maturity time T and strike price E .
The value of portfolio is = C + P
Straddle is effective when an investor is confident that a stock price will

change dramatically, but is uncertain of the direction of price move.
Short Straddle, = C P, profits when the underlying security

changes little in price before the expiration t = T .
Barings Bank was the oldest bank in London until its collapse in 1995. It
happened when the banks trader, Nick Leeson, took short straddle
positions and lost 1.3 billion dollars.

Bull Spread
Bull spread is a strategy that is designed to profit from a moderate rise in

the price of the underlying security.
Let us set up a portfolio consisting of a long position in call with strike

price E1 and short position in call with E2 such that E1 < E2 .
The value of this portfolio is t = Ct (E1 ) Ct (E2 ). At maturity t = T

0, S E1 ,
T = S E1 , E1 S < E2 ,
E2 E1 , S E2

The holder of this portfolio benefits when the stock price will be above
E1 .

Risk-Free Interest Rate
We assume the existence of risk-free investment. Examples: US

government bond, deposit in a sound bank. We denote by B(t) the value
of this investment.
Definition. Bond is a contact that yields a known amount F , called the

face value, on a known time T , called the maturity date. The authorized
issuer (for example, government) owes the holders a debt and is obliged to
pay interest (the coupon) and to repay the face value at maturity.
Zero-coupon bond involves only a single payment at T .

Return dB
B = rdt, where r is the risk-free interest rate.
If B(T ) = F , then B(t) = Fe r (T t) , where e r (T t) - discount factor

No Arbitrage Principle
The key principle of financial mathematics is No Arbitrage Principle.
There are never opportunities to make risk-free profit

Arbitrage opportunity arises when a zero initial investment 0 = 0 is
identified that guarantees non-negative payoff in the future such that
T > 0 with non-zero probability.
Arbitrage opportunities may exist in a real market. But, they cannot last
for a long time.

Lecture 6
1 No-Arbitrage Principle
2 Put-Call Parity
3 Upper and Lower Bounds on Call Options

No Arbitrage Principle
The key principle of financial mathematics is No Arbitrage Principle.
There are never opportunities to make risk-free profit.

identified that guarantees non-negative payoff in the future such that
Arbitrage opportunities may exist in a real market.
All risk-free portfolios must have the same return: risk-free interest rate.
Let be the value of a risk-free portfolio, and d is its increment during
a small period of time dt. Then
d
= rdt,

where r is the risk-less interest rate.
Let t be the value of the portfolio at time t. If T 0, then t 0
for t < T .
Put-Call Parity
Let us set up portfolio consisting of long one stock, long one put and short
one call with the same T and E .
The value of this portfolio is = S + P C .
The payoff for this portfolio is
T = S + max (E S, 0) max (S E , 0) = E
The payoff is always the same whatever the stock price is at t = T .
Using No Arbitrage Principle, we obtain
St + Pt Ct = Ee r (T t) ,
where Ct = C (St , t) and Pt = P (St , t).
This relationship between St , Pt and Ct is called Put-Call Parity which

represents an example of complete risk elimination.
Upper and Lower Bounds on Call Options
Put-Call Parity (t = 0): S0 + P0 C0 = Ee rT .
It shows that the value of European call option can be found from the
value of European put option with the same strike price and maturity:
C0 = P0 + S0 Ee rT .
Therefore C0 S0 Ee rT since P0 0 S0 Ee rT is the lower bound
for call option.
S0 Ee rT C0 S0
Let us illustrate these bounds geometrically.

Examples
Example 1. Find a lower bound for six months European call option with
the strike price 35 when the initial stock price is 40 and the risk-free
interest rate is 5% p.a.
In this case S0 = 40, E = 35, T = 0.5, and r = 0.05.
The lower bound for the call option price is S0 E exp (rT ) , or
40 35 exp (0.05 0.5) = 5.864
Example2. Consider the situation where the European call option is 4.

Show that there exists an arbitrage opportunity.
We establish a zero initial investment 0 = 0 by purchasing one call for

4 and the bond for 36 and selling one share for 40. The portfolio is
= C + B S.

Examples
At maturity t = T , the portfolio = C + B S has the value:

36.911 S, S 35
T = max (S E , 0) + 36 exp (0.05 0.5) S =
1.911 S > 35
It is clear that T > 0, therefore there exists an arbitrage opportunity.

Lecture 7
1 Upper and Lower Bounds on Put Options
2 Proof of Put-Call Parity by No-Arbitrage Principle
3 Example on Arbitrage Opportunity

Upper and Lower Bounds on Put Option
Reminder from lecture 6.

identified that guarantees a non-negative payoff in the future such that
Put-Call Parity at time t = 0: S0 + P0 C0 = Ee rT .
Upper and Lower Bounds on Put Option (exercise sheet 3):
Ee rT S0 P0 Ee rT
Let us illustrate these bounds geometrically.

Proof of Put-Call Parity
The value of European put option can be found as
P0 = C0 S0 + Ee rT .
Let us prove this relation by using No-Arbitrage Principle.
Assume that P0 > C0 S0 + Ee rT . Then one can make a riskless profit

(arbitrage opportunity).
We set up the portfolio = P S + C + B. At time t = 0 we
sell one put option for P0 (write the put option)
sell one share for S0 (short position)
buy one call option for C0
buy one bond for B0 = P0 + S0 C0 > Ee rT
The balance of all these transactions is zero, that is, 0 = 0

At maturity t = T the portfolio = P S + C + B has the value
(E S) S + B0 e rT , S E ,

T = = E + B0 e rT
S + (S E ) + B0 e rT , S > E ,
Since B0 > Ee rT , we conclude T > 0. and 0 = 0.
This is an arbitrage opportunity.

Now we assume that P0 < C0 S0 + Ee rT .
We set up the portfolio = P + S C B.
At time t = 0 we
buy one put option for P0
buy one share for S0 (long position)
sell one call option for C0 (write the call option)
borrow B0 = P0 + S0 C0 < Ee rT
The balance of all these transactions is zero, that is, 0 = 0
At maturity t = T we have T = E B0 e rT . Since B0 < Ee rT , we

conclude T > 0.
This is an arbitrage opportunity!!!

Example on Arbitrage Opportunity
Three months European call and put options with the exercise price 12
are trading at 3 and 6 respectively.
The stock price is 8 and interest rate is 5%. Show that there exists
arbitrage opportunity.
Solution:
The Put-Call Parity P0 = C0 S0 + Ee rT is violated, because

1
6 < 3 8 + 12e 0.05 4 = 6.851
To get arbitrage profit we

buy a put option for 6
sell a call option for 3
buy a share for 8
borrow 11 at the interest rate 5%.
The balance is zero!!
Example: Arbitrage Opportunity
1
The value of the portfolio = P + S C B at maturity T = 4 is
1
T = E B0 e rT = 12 11e 0.05 4 0.862.
1
Combination P + S C gives us 12. We repay the loan 11e 0.05 4 .
1
The balance 12 11e 0.05 4 is an arbitrage profit 0.862.

Lecture 8
1 One-Step Binomial Model for Option Price
2 Risk-Neutral Valuation
3 Examples

One-Step Binomial Model
Initial stock price is S0 . The stock price can either move up from S0 to
S0 u or down from S0 to S0 d ( u > 1; d < 1).
At time T , let the option price be Cu if the stock price moves up, and Cd
if the stock price moves down.
The purpose is to find the current price C0 of a European call option.

Riskless Portfolio
Now, we set up a portfolio consisting of a long position in shares and

short position in one call
= S C
Let us find the number of shares that makes the portfolio riskless.
The value of portfolio when stock moves up is
S0 u Cu
The value of portfolio when stock moves down is
S0 d Cd
If portfolio = S C is risk-free, then S0 u Cu = S0 d Cd

No-Arbitrage Argument
Cu Cd
The number of shares is = S0 (ud) .
Because portfolio is riskless for this , the current value 0 can be found
by discounting: 0 = (S0 u Cu ) e rT , where r is the interest rate.
On the other hand, the cost of setting up the portfolio is 0 = S0 C0 .

Therefore S0 C0 = (S0 u Cu ) e rT .
Finally, the current call option price is
C0 = S0 (S0 u Cu ) e rT ,
Cu Cd
where = S0 (ud) (No-Arbitrage Argument).

Risk-Neutral Valuation
Alternatively
C0 = e rT (pCu + (1 p)Cd ) ,
where
e rT d
p= .
ud
(Risk-Neutral Valuation)
It is natural to interpret the variable 0 p 1 as the probability of an up

movement in the stock price, and the variable 1 p as the probability of a
down movement.
Fair price of a call option C0 is equal to the expected value of its future
payoff discounted at the risk-free interest rate. For a put option P0 we
have the same result
P0 = e rT (pPu + (1 p)Pd ) .

Example
A stock price is currently $40. At the end of three months it will be either
$44 or $36. The risk-free interest rate is 12%.
What is the value of three-month European call option with a strike price
of $42? Use no-arbitrage arguments and risk-neutral valuation.
In this case S0 = 40, u = 1.1, d = 0.9, r = 0.12, T = 0.25, Cu = 2,

Cd = 0.
No-arbitrage arguments: the number of shares

Cu Cd 20
= = = 0.25
S0 u S0 d 40 (1.1 0.9)
and the value of call option
C0 = S0 (S0 u Cu ) e rT =
40 0.25 (40 1.1 0.25 2) e 0.120.25 = 1.266

Example
Risk-neutral valuation: one can find the probability p
e rT d e 0.120.25 0.9
p= = = 0.6523
ud 1.1 0.9
and the value of call option
C0 = e rT [pCu + (1 p)Cd ] = e 0.120.25 [0.6523 2 + 0] = 1.266

Lecture 9
1 Risk-Neutral Valuation
2 Risk-Neutral World
3 Two-Steps Binomial Tree

Risk-Neutral Valuation
Reminder from Lecture 8. Call option price:
C0 = e rT (pCu + (1 p)Cd ) ,
e rT d
where p = ud . No-Arbitrage Principle: d < e rT < u.
In particular, if d > e rT then there exists an arbitrage opportunity. We

could make money by taking out a bank loan B0 = S0 at time t = 0 and
buying the stock for S0 .
We interpret the variable 0 p 1 as the probability of an up movement

in the stock price.
This formula is known as a risk-neutral valuation.
The probability of up q or down movement 1 q in the stock price plays

no role whatsoever! Why???
Risk Neutral Valuation
Let us find the expected stock price at t = T :

e rT d e rT d
Ep [ST ] = pS0 u + (1 p)S0 d = ud S0 u + (1 ud )S0 d = S0 e rT .
This shows that stock price grows on average at the risk-free interest rate
r . Since the expected return is r , this is a risk-neutral world.
In Real World: E [ST ] = S0 e T . In Risk-Neutral World: Ep [ST ] = S0 e rT
Risk-Neutral Valuation: C0 = e rT Ep [CT ]
The option price is the expected payoff in a risk-neutral world, discounted

at risk-free rate r .

Two-Step Binomial Tree
Now the stock price changes twice, each time by either a factor of u > 1
or d < 1. We assume that the length of the time step is t such that
T = 2t. After two time steps the stock price will be S0 u 2 , S0 ud or S0 d 2 .
The call option expires after two time steps producing payoffs Cuu , Cud
and Cdd respectively.
Option Price
The purpose is to calculate the option price C0 at the initial node of the
tree.
We apply the risk-neutral valuation backward in time:
Cu = e r t (pCuu + (1 p)Cud ) , Cd = e r t (pCud + (1 p)Cdd ) .
Current option price: C0 = e r t (pCu + (1 p)Cd ) .

Option Price
Substitution gives
C0 = e 2r t p 2 Cuu + 2p(1 p)Cud + (1 p)2 Cdd ,

where p 2 , 2p(1 p) and (1 p)2 are the probabilities in a risk-neutral

world that the upper, middle, and lower final nodes are reached.
Finally, the current call option price is
C0 = e rT Ep [CT ] , T = 2t.
The current put option price can be found in the same way:
P0 = e 2r t p 2 Puu + 2p(1 p)Pud + (1 p)2 Pdd

or
P0 = e rT Ep [PT ] .

Two-Step Binomial Tree Example
Consider six months European put with a strike price of 32 on a stock
with current price 40. There are two time steps and in each time step
the stock price either moves up by 20% or moves down by 20%. Risk-free
interest rate is 10%. Find the current option price.
We have u = 1.2, d = 0.8, t = 0.25, and r = 0.1.

rt d 0.10.25 0.8
Risk-neutral probability p = e ud = e 1.20.8 = 0.5633.
The possible stock prices at final nodes are 40 (1.2)2 = 57.6,

40 1.2 0.8 = 38.4, and 40 (0.8)2 = 25.6.
We obtain Puu = 0, Pud = 0 and Pdd = 32 25.6 = 6.4.
Thus the value of put option is

P0 = e 20.10.25 0 + 0 + (1 0.5633)2 6.4 = 1.1610.

Lecture 10
1 Binomial Model for Stock Price
2 Option Pricing on Binomial Tree
3 Matching Volatility with u and d

Binomial model for the stock price
Continuous random model for the stock price: dS = Sdt + SdW
The binomial model for the stock price is a discrete time model:
The stock price S changes only at discrete times t, 2t, 3t, ...
The price either moves up S Su or down S Sd with d < e r t < u.
The probability of up movement is q.
Let us build up a tree of possible stock prices. The tree is called a binomial
tree, because the stock price will either move up or down at the end of
each time period. Each node represents a possible future stock price.
We divide the time to expiration T into several time steps of duration

t = T /N, where N is the number of time steps in the tree.
Example: Let us sketch the binomial tree for N = 4.

Stock Price Movement in the Binomial Model
We introduce the following notations:
Snm is the n-th possible value of stock price at time-step mt.
Then Snm = u n d mn S00 , where n = 0, 1, 2, ..., m.
S00 is the stock price at the time t = 0. Note that u and d are the same at
every node in the tree.
For example, at the third time-step 3t, there are four possible stock
prices: S03 = d 3 S00 , S13 = ud 2 S00 , S23 = u 2 dS00 and S33 = u 3 S00 .
At the final time-step Nt, there are N + 1 possible values of stock price.

Call Option Pricing on Binomial Tree
We denote by Cnm the n-th possible value of call option at time-step mt.
Risk Neutral Valuation (backward in time):
Cnm = e r t pCn+1
m+1
+ (1 p)Cnm+1 .

e rt d
Here 0 n m and p = ud .
Final condition: CnN = max SnN E , 0 , where n = 0, 1, 2, ..., N, E is the

strike price.
The current option price C00 is the expected payoff in a risk-neutral world,
discounted at risk-free rate r : C00 = e rT Ep [CT ] .
Example: N = 4.

Matching volatility with u and d
We assume that the stock price starts at the value S0 and the time step is
t. Let us findthe expected stock price, E [S] , and the variance of the
return, var SS , for continuous and discrete models.
Expected stock price: Continuous model: E [S] = S0 e t .

On the binomial tree: E [S] = qS0 u + (1 q)S0 d.
First equation: qu + (1 q)d = e t .
Variance of the return: Continuous model: var S = 2 t (Lecture2)

S S
On the binomial tree: var S =
q(u 1)2 + (1 q)(d 1)2 [q(u 1) + (1 q)(d 1)]2 =
qu 2 + (1 q)d 2 [qu + (1 q)d]2 .
Recall: var [X ] = E X 2 [E (X )]2 .

Second equation: qu 2 + (1 q)d 2 [qu + (1 q)d]2 = 2 t.

Third equation: u = d 1 .
Matching volatility with u and d
e t d
From the first equation we find q = ud .
This is the probability of an up movement in the real world.
Substituting this probability into the second equation, we obtain

e t (u + d) ud e 2t = 2 t.
Using u = d 1 , we get
1
e t (u + ) 1 e 2t = 2 t.
u
This equation can be reduced to the quadratic equation. (Exercise sheet 4,
part 5).
One can obtain u e t 1 + t and d e t .
These are the values of u and d obtained by Cox, Ross, and Rubinstein in
1979.
Recall: e x 1 + x for small x.

Lecture 11
1 American Put Option Pricing on Binomial Tree
2 Replicating Portfolio

American Option
An American Option is one that may be exercised at any time prior to

expire (t = T ).
We should determine when it is best to exercise the option.
It is not subjective! It can be determined in a systematic way!
The American put option value must be greater than or equal to the
payoff function.
If P < max(E S, 0), then there is obvious arbitrage opportunity.
We can buy stock for S and option for P and immediately exercise the
option by selling stock for E .
E (P + S) > 0

American Put Option Pricing on Binomial Tree
We denote by Pnm the n-th possible value of put option at time-step mt.
European Put Option:
Pnm = e r t pPn+1
m+1
+ (1 p)Pnm+1 .

e rt d
Here 0 n m and the risk-neutral probability p = ud .
American Put Option:

n o
Pnm = max max(E Snm , 0), e r t pPn+1
m+1
+ (1 p)Pnm+1 ,
where Snm is the n-th possible value of stock price at time-step mt.
Final condition: PnN = max E SnN , 0 , where n = 0, 1, 2, ..., N, E is the

strike price.
Example: Evaluation of American Put Option on Two-Step
Tree
We assume that over each of the next two years the stock price either
moves up by 20% or moves down by 20%. The risk-free interest rate is 5%.
Find the value of a 2-year American put with a strike price of $52 on a
stock whose current price is $50.
In this case u = 1.2, d = 0.8, r = 0.05, E = 52.

e 0.05 0.8
Risk-neutral probability: p= 1.20.8 = 0.6282

Replicating Portfolio
The aim is to calculate the value of call option C0 .
Let us establish a portfolio of stocks and bonds in such a way that the
payoff of a call option is completely replicated.
Final value: T = CT = max (S E , 0)
To prevent risk-free arbitrage opportunity, the current values should be

identical. We say that the portfolio replicates the option.
The Law of One Price: t = Ct .
Consider replicating portfolio of shares held long and N bonds held

short.
The value of portfolio: = S NB. A pair (, N) is called a trading
strategy.
How to find (, N) such that T = CT and 0 = C0 ?

Example: One-Step Binomial Model.
Initial stock price is S0 . The stock price can either move up from S0 to
S0 u or down from S0 to S0 d. At time T , let the option price be Cu if the
stock price moves up, and Cd if the stock price moves down.
The value of portfolio: = S NB.
When stock moves up: S0 u NB0 e rT = Cu .
When stock moves down: S0 d NB0 e rT = Cd .
We have two equations for two unknown variables and N.
Current value: C0 = S0 NB0 .

e rT d
Prove: C0 = e rT (pCu + (1 p)Cd ) , where p = ud . (Exercise sheet 5)

Lecture 12
1 Black-Scholes Model
2 Black-Scholes Equation
The Black-Scholes model for option pricing was developed by Fischer

Black, Myron Scholes in the early 1970s. This model is the most
important result in financial mathematics.

Black - Scholes model
The Black-Scholes model is used to calculate an option price using: stock
price S, strike price E , volatility , time to expiration T , and risk-free
interest rate r .
This model involves the following explicit assumptions:
The stock price follows a Geometric Brownian motion with constant

expected return and volatility: dS = Sdt + SdW . .
No transaction costs.
The stock does not pay dividends.
One can borrow and lend cash at a constant risk-free interest rate.
Securities are perfectly divisible (i.e. one can buy any fraction of a share
of stock).
No restrictions on short selling.

Basic Notation
We denote by V (S, t) the value of an option. We use the notations

C (S, t) and P(S, t) for call and put when the distinction is important.
The aim is to derive the famous Black-Scholes Equation:
V 1 2V V
+ 2 S 2 2 + rS rV = 0.
t 2 S S
Now, we set up a portfolio consisting of a long position in one option and
a short position in shares.
The value is = V S.
Let us find the number of shares that makes this portfolio riskless.

Itos Lemma and Elimination of Risk
The change in the value of this portfolio in the time interval dt:
d = dV dS, where dS = Sdt + SdW .
Using Itos Lemma:
1 2 2 2V

V V V
dV = + S 2
+ S dt + S dW ,
t 2 S S S
we find
1 2 2 2V

V V V
d = + S + S S dt + S S dW .
t 2 S 2 S S
The main question is how to eliminate the risk!!!

V
We can eliminate the random component in d by choosing = S .

Black-Scholes Equation
This choice
results in a risk-free
portfolio = V S V
S whose increment
V 1 2 2 V 2
is d = t + 2 S S 2 dt.
No-Arbitrage Principle: the return from this portfolio must be rdt.

2V
V
+ 12 2 S 2 V 2
d S 2
+ 12 2 S 2 SV2 = r V S V
t

= rdt or V S V
= r or t S .
S
Thus, we obtain the Black-Scholes PDE:
V 1 2V V
+ 2 S 2 2 + rS rV = 0.
t 2 S S
Scholes received the 1997 Nobel Prize in Economics. It was not awarded
to Black in 1997, because he died in 1995.
Black received a Ph.D. in applied mathematics from Harvard University.

The European call value C (S, t)
If PDE is of backward type, we must impose a final condition at t = T .

For a call option, we have C (S, T ) = max(S E , 0).

Lecture 13
1 Boundary Conditions for Call and Put Options
2 Exact Solution to Black-Scholes Equation
C 1 2C C
+ 2 S 2 2 + rS rC = 0.
t 2 S S

Boundary Conditions
We use C (S, t) and P(S, t) for call and put option. Boundary conditions
are applied for zero stock price S = 0 and S .
Boundary conditions for a call option:
C (0, t) = 0 and C (S, t) S as S .
The call option is likely to be exercised as S
Boundary conditions for a put option:
P(0, t) = Ee r (T t) . We evaluate the present value of E .
P(S, t) 0 as S .
As stock price S , then put option is unlikely to be exercised.

Exact solution
The Black-Scholes equation
C 1 2C C
+ 2 S 2 2 + rS rC = 0
t 2 S S
with appropriate final and boundary conditions has the explicit solution:
C (S, t) = SN (d1 ) Ee r (T t) N (d2 ) ,
where
x
1
Z
y2
N (x) = e 2 dy (cumulative normal distribution)
2
and
ln (S/E ) + r + 2 /2 (T t)

d1 = , d2 = d1 T t.
T t

The European call value C (S, t) by K. Rubash

Example
Calculate the price of a three-month European call option on a stock with
a strike price of $25 when the current stock price is $21.6 The volatility is
35% and risk-free interest rate is 1% p.a.
In this case S0 = 21.6, E = 25, T = 0.25, = 0.35 and r = 0.01.
The value of call option is C0 = S0 N (d1 ) Ee rT N (d2 ) .
First, we compute the values of d1 and d2 :
ln(S0 /E )+(r +2 /2)T ln( 21.6
25 )
+(0.01+(0.35)2 0.5))0.25
d1 =
T
=
0.35 0.25
0.7335

d2 = d1 T = 0.7335 0.35 0.25 0.9085
Since
N (0.7335) 0.2316, N (0.9085) 0.1818,
we obtain
C0 21.6 0.2316 25 e 0.010.25 0.1818 = 0.4689
The Limit of Higher Volatility
Let us find the limit
lim C (S, t).

We know that
C (S, t) = SN (d1 ) Ee r (T t) N (d2 ) ,

ln(S/E )+(r +2 /2)(T t) ln(S/E ) r T t T t
where d1 =
T t
=
T t
+ + 2 .
In the limit , d1 .

Since d2 = d1 T t, in the limit , d2
Thus lim N (d1 ) = 1 and lim N(d2 ) = 0.
Therefore lim C (S, t) = S. This is an upper bound for call option!!!

Lecture 14
1 -Hedging
2 Greek Letters or Greeks

Delta for European Call Option
C
Let us show that = S = N (d1 ) .
First, find the derivative = C

S by using the explicit solution for the
European call C (S, t) = SN (d1 ) Ee r (T t) N (d2 ).
C d1 d2
= = N (d1 ) + SN (d1 ) Ee r (T t) N (d2 ) =
S S S
d1
N (d1 ) + (SN (d1 ) Ee r (T t) N (d2 )) .
S
We need to prove
(SN (d1 ) Ee r (T t) N (d2 )) = 0.
See Problem Sheet 6.

Delta-Hedging Example
Find the value of of a 6-month European call option on a stock with a

strike price equal to the current stock price ( t = 0). The interest rate is
6% p.a. The volatility = 0.16.
ln(S0 /E )+(r +2 /2)T

Solution: we have = N (d1 ) , where d1 = 1
(T ) 2
We find
(0.06 + (0.16)2 0.5)) 0.5

d1 = 0.3217
0.16 0.5
and
= N (0.3217) 0.6262

Delta for European Put Option
Let us find Delta for European put option by using the put-call parity:
S + P C = Ee r (T t) .
Let us differentiate it with respect to S

P C
We find 1 + S S = 0.
P C
Therefore S = S 1 = N (d1 ) 1.

Greeks
The option value: V = V (S, t | , r , T ).
Greeks represent the sensitivities of options to a change in underlying

parameters on which the value of an option is dependent.
Delta:
= V
S measures the rate of change of option value with respect to
changes in the underlying stock price
Gamma:
2
= SV2 =
S measures the rate of change in with respect to changes
in the underlying stock price.
d12
2C (d )
N 1 1 e
Problem Sheet 6: = S 2
= S T t
, where N (d1 ) = 2
2 .

Greeks
V
Vega, , measures sensitivity to volatility .
C

One can show that = S T tN (d1 ).
Rho:
V
= r measures sensitivity to the interest rate r .
C
One can show that = r = E (T t)e r (T t) N(d2 ).
The Greeks are important tools in financial risk management. Each Greek
measures the sensitivity of the value of derivatives or a portfolio to a small
change in a given underlying parameter.

Lecture 15
1 Black-Scholes Equation and Replicating Portfolio
2 Static and Dynamic Risk-Free Portfolio

Replicating Portfolio
The aim is to show that the option price V (S, t) satisfies the
Black-Scholes equation
V 1 2V V
+ 2 S 2 2 + rS rV = 0.
t 2 S S
Consider replicating portfolio of shares held long and N bonds held
short. The value of portfolio: = S NB. Recall that a pair (, N) is
called a trading strategy.
How to find (, N) such that t = Vt ?
SDE for a stock price S(t) : dS = Sdt + SdW .
Equation for a bond price B (t) : dB = rBdt.

Derivation of the Black-Scholes Equation
By using the Itos lemma, we find the change in the option value
1 2 2 2V

V V V
dV = + S + S 2
dt + S dW .
t S 2 S S
By using self-financing requirement d = dS NdB, we find the change
in portfolio value
d = dSNdB = (Sdt+SdW )rNBdt = (SrNB)dt+SdW
Equating the last two equations d = dV , we obtain

V V 2
= S , rNB = t + 12 2 S 2 SV2 .
S V

Since NB = S = S V , we get the classical Black-Scholes
equation
V 1 2 2 2V V
+ S 2
+ rS rV = 0.
t 2 S S
Static Risk-free Portfolio
Let me remind you a Put-Call Parity. We set up the portfolio consisting of

long position in one stock, long position in one put and short position in
one call with the same T and E .
The value of this portfolio is = S + P C .
The payoff for this portfolio is
T = S + max (E S, 0) max (S E , 0) = E
The payoff is always the same whatever the stock price is at t = T .
Using No Arbitrage Principle, we obtain

St + Pt Ct = Ee r (T t) ,
where Ct = C (St , t) and Pt = P (St , t).
This is an example of complete risk elimination.
Definition: The risk of a portfolio is the variance of the return.

Dynamic Risk-Free Portfolio
Put-Call Parity is an example of complete risk elimination when we carry
out only one transaction in call/put options and underlying security.
Let us consider the dynamic risk elimination procedure.
We could set up a portfolio consisting of a long position in one call option

and a short position in shares.
The value is = C S.
We can eliminate the random component in by choosing

C
= .
S
This is a -hedging! It requires a continuous rebalancing of a number of

shares in the portfolio .
Lecture 16
1 Option on Dividend-paying Stock
2 American Put Option

Option on Dividend-paying Stock
We assume that in a time dt the underlying stock pays out a dividend
D0 Sdt,
where D0 is a constant dividend yield.
Now, we set up a portfolio consisting of a long position in one call option

and a short position in shares.
The value is = C S.
d = dC dS D0 Sdt.

Using Itos Lemma:
2C

C 1 C
dC = + 2 S 2 2 dt + dS
t 2 S S
we find
2C

C 1 C C
d = + 2 S 2 2 D0 S dt + dS dS
t 2 S S S
C

Modified Black-Scholes Equation
This choice
results in a 2risk-free portfolio
= C S C
S whose increment
C 1 2 2 C C
is d = t + 2 S S 2 D0 S S dt.
No-Arbitrage Principle: the return from this portfolio must be rdt.

C 2
d
+ 12 2 S 2 SC2 D0 S C C

= rdt or t S = r C S S .
Thus, we obtain the modified Black-Scholes PDE:
C 1 2C C
+ 2 S 2 2 + (r D0 )S rC = 0.
t 2 S S

Solution to Modified Black-Scholes Equation
Let us find the solution to modified Black-Scholes equation in the form
C (S, t) = e D0 (T t) C1 (S, t).
We prove that C1 (S, t) satisfies the Black-Scholes equation with r

replaced by r D0 .
If we substitute C (S, t) = e D0 (T t) C1 (S, t) into the modified

Black-Scholes equation, we find the equation for C1 (S, t) in the form
C1 1 2 2 2 C1 C1
+ S 2
+ (r D0 ) (r D0 )C1 = 0
t 2 S S
. The auxiliary function C1 (S, t) is the value of a European Call option
with the interest rate r D0 .

Solution to Modified Black-Scholes Equation
Problem sheet 7: show that the modified Black-Scholes equation has the
explicit solution for the European call
C (S, t) = Se D0 (T t) N (d10 ) Ee r (T t) N (d20 ) ,
where
ln (S/E ) + r D0 + 2 /2 (T t)

d10 = , d20 = d10 T t.
T t

American Put Option
Recall that An American Option is one that may be exercised at any time
prior to expire (t = T ).
The American put option value must be greater than or equal to the
payoff function.
If P < max(E S, 0), then there is obvious arbitrage opportunity.

American Put Option
American put problem can be written as as a free boundary problem.
We divide the price axis S into two distinct regions:

0 S < Sf (t) and Sf (t) < S < ,
where Sf (t) is the exercise boundary. Note that we do not know a priori
the value of Sf (t).
When 0 S < Sf (t), the early exercise is optimal: put option value is
P(S, t) = E S.
When S > Sf (t), the early exercise is not optimal, and P(S, t) obeys the
Black-Scholes equation.
The boundary conditions at S = Sf (t) are
P
P (Sf (t), t) = max (E Sf (t), 0) , (Sf (t), t) = 1.
S
Lecture 17
1 Bond Pricing with Known Interest Rates and Dividend Payments
2 Zero-Coupon Bond Pricing

Bond Pricing Equation
Definition. A bond is a contract that yields a known amount (nominal,

principal or face value) on the maturity date, t = T . The bond pays the
dividends at fixed times.
If there is no dividend payment, the bond is known as a zero-coupon bond.
Let us introduce the following notation
V (t) is the value of the bond, r (t) is the interest rate, K (t) is the
coupon payment.
Equation for the bond price is

dV
= r (t)V K (t).
dt
The final condition: V (T ) = F .
Zero-Coupon Bond Pricing
Let us consider the case when the dividend payment K (t) = 0.

dV
The solution of the equation dt = r (t)V with V (T ) = F can be written
as Z T
V (t) = F exp r (s)ds .
t
dV
Let us show this by integration V = r (t)dt from t to T .
RT RT
ln V (T ) ln V (t) = t r (s)ds or ln(V (t)/F ) = t r (s)ds.

Example
A zero coupon bond, V , issued at t = 0, is worth V (1) = 1 at t = 1. Find
the bond price V (t) at time t < 1 and V (0), when the continuous interest
rate is
r (t) = t 2 .
Solution: T = 1; one can find

Z 1 1
1 1 3
Z
r (s)ds = s 2 ds = t
t t 3 3
Therefore
Z 1 3
t 1
V (t) = exp r (s)ds = exp ,
t 3 3

1
V (0) = exp = 0.7165
3

Bond Pricing for K (t) > 0
Let us consider the case when the dividend payment K (t) > 0. The
solution of the equation dV
dt = r (t)V K (t) can be written as
Z T
V (t) = F exp r (s)ds + V1 (t) ,
t
R
T
where V1 (t) = C (t) exp t r (s)ds .
To find C (t) we need to substitute V1 (t) into the equation

dV1
= r (t)V1 K (t).
dt
(tutorial exercise 8)
The explicit solution is

Z T Z T Z T
V (t) = exp r (s)ds F+ K (y ) exp r (s)ds dy .
t t y

Lecture 18
1 Measure of Future Values of Interest Rate
2 Term Structure of Interest Rate (Yield Curve)

Measure of Future Value of Interest Rate
Let us consider the case when the dividend payment K (t) = 0. Recall that
the solution of the equation dV
dt = r (t)V with V (T ) = F is
Z T
V (t) = F exp r (s)ds .
t
Let us introduce the notation V (t, T ) for the bond prices. These prices
are quoted at time t for different values of T .
Let us differentiate V (t, T ) with respect to T :

R
V T
T = F exp t r (s)ds (r (T )) = V (t, T )r (T ),

RT
since T t r (s)ds = r (T ). Therefore,
1 V
r (T ) = .
V (t, T ) T
This is an interest rate at future dates (forward rate).
Term Structure of Interest Rate (Yield Curve)
We define
ln(V (t, T )) ln V (T , T )
Y (t, T ) =
T t
as a measure of the future values of interest rate, where V (t, T ) is taken
from financial data.
R
ln(F exp tT r (s)ds )ln F RT
1
Y (t, T ) = T t = T t t r (s)ds
is the average value of the interest rate r (t) in the time interval [t, T ].
Bond price V (t, T ) can be written as V (t, T ) = Fe Y (t,T )(T t) .
Term structure of interest rate (yield curve):

T
ln(V (0, T )) ln V (T , T ) 1
Z
Y (0, T ) = = r (s)ds
T T 0
is the average value of interest rate in the future.

Example
Assume that the instantaneous interest rate r (t) is
r (t) = r0 + at,
where r0 and a are constants.
Bond price:
RT RT a
= Fe r0 (T t) 2 (T ).
r (s)ds (r0 +as)ds 2 t 2
V (t, T ) = Fe t = Fe t
Term structure of interest rate:

T
1
Z
aT
Y (0, T ) = r (s)ds = r0 + .
T 0 2

Risk of Default
There exists a risk of default of bond V (t) when the principal are not paid
to the lender as promised by borrower.
How to take this into account?
Let us introduce the probability of default p during one year T = 1. Then
V (0) e r = V (0) (1 p)e (r +s) + p 0.
The investor has no preference between two investments.

The positive parameter s is called the spread w.r.t to interest rate r .
Let us find it.

s = ln(1 p) = p + o(p).

Lecture 19
1 Asian Options
2 Derivation of PDE for Option Price

Asian Options
Definition. Asian option is a contract giving the holder the right to
buy/sell an underlying asset for its average price over some prescribed
period.
Asian call option final condition:

T
1
Z
V (S, T ) = max S S(t)dt, 0 .
T 0
We introduce a new variable:
Z t
dI
I (t) = S(t)dt or = S(t).
0 dt
Final condition can be rewritten as

I
V (S, I , T ) = max S , 0 .
T

Asian Options
Asian option price V (S, I , t) is the function of three variables.
Let us derive the equation for this price by using the standard portfolio
consisting of a long position in one call option and a short position in
shares.
The value is = V S.
d = dV dS.

Using Itos Lemma:
1 2 2 2V

V V V
dV = + S 2
dt + dS + dI
t 2 S S I
we find
2V

V 1 V V
d = + 2 S 2 2 dt + dS dS + dI .
t 2 S S I
V

Modified Black-Scholes Equation
By using No-Arbitrage Principle, and the equation
dI = Sdt
we can obtain the modified Black-Scholes PDE for the Asian option price:
V 1 2V V V
+ 2 S 2 2 + rS rV + S = 0.
t 2 S S I
The value of Asian option must be calculated numerically (no analytical

solution!)

20912lectures01 19

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

20912lectures01 19

Uploaded by

Copyright:

Available Formats

Lectures

20912 - Introduction to Financial Mathematics

No tutorials in the first week

Sergei Fedotov (University of Manchester) 20912 2010 1/1

2 Time Value of Money

3 Continuous Model for Stock Price

Sergei Fedotov (University of Manchester) 20912 2010 2/1

J. Hull, Options, Futures and Other Derivatives, 7th Edition,

Test in a week 6: 20%

Sergei Fedotov (University of Manchester) 20912 2010 3/1

Stock Markets, such as NYSE(New York Stock Exchange), London

Bond Markets, where participants buy and sell debt securities.

Option market is massive! More money is invested in options than in the

Why stochastic differential equations (SDEs) and partial differential

Return (relative measure of change):

How to model the return?

Sergei Fedotov (University of Manchester) 20912 2010 6/1

dW describes the stochastic change in the stock price, where dW

W (t) is a Wiener process

is the volatility ( = 0.2 0.5)

dS = Sdt + SdW (7)

Simple case: volatility = 0

Sergei Fedotov (University of Manchester) 20912 2010 8/1

W (t) has independent increments: if u v s t, then W (t) W (s)

W (s + t) W (s) is N(0, t) and W (0) = 0

The increment W = W (t + t) W (t) can be written as

1 Properties of Wiener Process

2 Approximation for Stock Price Equation

Sergei Fedotov (University of Manchester) 20912 2010 10 / 1

Simulations of a Wiener process:

Sergei Fedotov (University of Manchester) 20912 2010 11 / 1

The increment W = W (t + t) W (t) can be written as

Sergei Fedotov (University of Manchester) 20912 2010 12 / 1

Answer: S = 0.288 + 4.155X

Note: 4.155 is the standard deviation

Sergei Fedotov (University of Manchester) 20912 2010 13 / 1

Find df if dS = Sdt + SdW

Example. Find the SDE satisfied by f = S 2 .

Sergei Fedotov (University of Manchester) 20912 2010 14 / 1

1 Distribution for ln S(t)

2 Solution to Stochastic Differential Equation for Stock Price

Sergei Fedotov (University of Manchester) 20912 2010 15 / 1

by using Itos Lemma:

Integration from 0 to t gives

We obtain for ln S(t)

Example 2. Consider a stock with an initial price of 40, an expected return

Answer: ln S(0.5) N (3.759, 0.020)

The probability density function of X = ln S(t) is

Definition. The model of a stock dS = Sdt + SdW is known as a

The random function S(t) can be found from

Sergei Fedotov (University of Manchester) 20912 2010 19 / 1

2 European Call and Put Options

3 Payoff Diagrams, Short Selling and Profit

Sergei Fedotov (University of Manchester) 20912 2010 20 / 1

The question is what does this actually mean?

Consider a three-month European call option on a BP share with a strike

Sergei Fedotov (University of Manchester) 20912 2010 22 / 1

Definition. Payoff Diagram ia a graph of the value of the option position

Sergei Fedotov (University of Manchester) 20912 2010 23 / 1

Definition. Portfolio is the combination of assets, options and bonds.

We denote by the value of portfolio. Example: = 2S + 4C 5P.

It means that portfolio consists of long position in two shares, long

Sergei Fedotov (University of Manchester) 20912 2010 25 / 1