Mathematical Finance

Lecture Note1
1. A single-period model

Goal: introduce a basic model for a nancial

market, which will be further developed and
extended throughout the course.
1.1 The market model

Model assumptions:

• Time scale: consider market at only two

instants: t = 0 and t = 1 in this chapter.
This is called a single-period model as
there is just one period between t = 0 and
t = 1.

• Assets: We consider just two assets:

  a risky asset: the value at time 1 is
random and unknown at time 0; typical
example: a stock, but can also be a
foreign currency or a commodity.
 a risk-free asset: the value at time 1
is deterministic and known at time 0;
think of a bank account or a bond of an
issuer with excellent credit rating;

• Notation:
 S(t) is the price of the risky asset at
time t; S(0) is a positive constant while
S(1) is a non-negative random variable.

 A(t) is the price of the risk-free asset at

time t; both A(0) and A(1) are positive
constants;

• A portfolio (x, y) consists of two numbers

x and y , which represent the units of risky
 and risk-free asset, respectively, which the
investor decides to invest at time t = 0.
We assume that x and y can take any value
in R.

• The value of a portfolio (x, y) is denoted

by V (t) for t = 0, 1 and given by
V (t) = xS(t) + yA(t), t = 0, 1.
The change in the portfolio value equals
   
V (1)−V (0) = x S(1)−S(0) +y A(1)−A(0) .

• A portfolio (x, y) is said to be admissible

if its value is nonnegative at both times:
V (0) > 0 and V (1) > 0.
1.2 No arbitrage

The no-arbitrage principle (which essentially says an

investor can not start with zero portfolio and have a possibility
) is the most fundamental
of earning positive money
assumption about a reasonable model for a -
nancial market.
Denitions:

• An admissible portfolio (x, y) is an arbi-

trage opportunity if its value process sat-
ises
and P
 
V (0) = 0 V (1) > 0 > 0

• This means you can start with zero invest-

ment and end up with a value, which is
always nonnegative (admissible portfolio)
and positive in some scenario.
 • In reality, arbitrage opportunities rarely ex-
ist, and when they do, they disappear quickly
as traders would immediately benet from
them, making the market free of arbitrage
opportunities. Therefore, it is reasonable
to assume that the model has no arbitrage
opportunities.

• We say that the market model is arbitrage

free if there do not exist any arbitrage op-
portunities.

Example

• Consider a model where S(1) can take just

two values,
 S u with probability p


S(1) =
 Sd with probability 1 − p
 where S d < S u and p ∈ (0, 1).

• This is called the one-step binomial model.

It corresponds to two scenarios: an upward
and a downward stock price.

Proposition 1.1 Suppose that S(0) = A(0).

Then the one-step binomial model is arbitrage
free if and only if
S d < A(1) < S u. 1.1
Proof: Let (x, y) be an admissible portfolio
such that V (0) = 0. Then

V (0) = 0 =⇒ xS(0) + yA(0) = 0 =⇒ x = −y.

Therefore, we have
 
V (1) = xS(1) + yA(1) = x S(1) − A(1)
   
 x S u − A(1) > 0
=  
 x S d − A(1) > 0

where we used for the last inequalities the ad-

missibility condition.
If (1.1) holds, we have
x > 0 and x 6 0 =⇒ x = 0 =⇒ V (1) = 0
=⇒ (x, y) is not an arbitrage opportunity.
=⇒ The model is arbitrage free.

Conversely, if (1.1) does not hold, we have

either S d > A(1) or A(1) > S u.
Using S d < S u by assumption, the
case S d > A(1) implies S u > S d > A(1)

=⇒ V (1) > 0 and P V (1) > 0 > 0 for all x > 0

 

=⇒ there exist arbitrage opportunities.

In the case A(1) > Su, we have A(1) > S u > S d

=⇒ V (1) > 0 and P V (1) > 0 > 0 for all x < 0
=⇒ there exist arbitrage opportunities.
Remark: The proof also shows how we can
nd an arbitrage opportunity if (1.1) does not
hold.

• In the case S d > A(1), we have

S d > A(1) and S u > A(1)
=⇒ stock performs better than risk-free
asset
=⇒ buy stock and borrow from risk-free
asset
=⇒ (x, −x) with x > 0 is an arbitrage op-
portunity.

• In the case S u 6 A(1), we have

S d < A(1) and S u 6 A(1)
=⇒ stock performs worse than risk-free as-
set
=⇒ sell stock and invest in risk-free asset
=⇒ (x, −x) with x < 0 is an arbitrage op-
portunity.
Example 1) Assume that A(0) = A(1) = 100
and S(0) = 100
with prob. 0.25

110
S(1) =
105 with prob. 0.75

a) Find an arbitrage opportunity which gives a

risk-free prot of at least 500 in either case.
Solution:
Consider a portfolio (x, y) with V (0) = 0.
From V (0) = 100x + 100y = 0, it follows y =
−x and hence
10x with prob. 0.25


V (1) = S(1)x−100x =
5x with prob. 0.75
so we need x > 100 to get at least 500 in either
case.
b) Find an arbitrage opportunity which gives
an expected risk-free prot of 75.
From 10x 1
4 + 5x 3 = 75,
4 we get x = 12 and
y = −12.

Example 2) As in 1), but with A(1) = 110.

Can we nd portfolios for questions a) and b)?
Solution:
From V (0) = 0, we still have y = −x, but now
with prob.

 0 0.25
V (1) = S(1)x−110x =
−5x with prob. 0.75

For a), it is not possible to get 500 in both

cases.
For b),we nd from 0 41 − 5x 34 = 75 that
x = −20 and y = 20.
Example 3) Assume that A(1) = A(0) = 100,
S(0) = 80 and
with probability 0.25

 110

S(1) = 105 with probability 0.5


with probability 0.25




50
Is the model arbitrage free? If not, give an
arbitrage opportunity.
Solution:
We consider again a portfolio (x, y) with
V (0) = 0, which implies

V (0) = 80x + 100y = 0,

so y = − 54 x. This implies
with prob. 0.25

 30x

25x with prob. 0.5

V (1) = S(1)x−80x =
−30x with prob. 0.25




The condition V (1) > 0 yields x > 0 and x 6 0,

hence x = 0 and y = 0 so that V (1) = 0.
Therefore, the model is arbitrage free.
Example 4) Assume that A(1) = A(0) = 100,
S(0) = 80 and
with probability 0.25

 110

with probability 0.5

S(1) = s
with probability 0.25



t
where s and t are nonnegative numbers. Give
necessary and sucient conditions on s and t
such that the model is free from arbitrage.
Solution:
As in Example 3), we have y = − 54 x for a port-
folio (x, y) with V (0) = 0. We obtain
with prob. 0.25

 30x

V (1) = S(1)x−80x = (s − 80)x with prob. 0.5


(t − 80)x with prob. 0.25





The condition V (1) > 0 yields x > 0 from the

rst case. For t, s > 0, we can conclude
arbitrage ⇐⇒ t > 80 and s > 80,
arbitrage free ⇐⇒ t < 80 or s < 80.
1.3 Risk and return

For the assets given in Section 1.1, we can

dene and calculate their returns.
Denitions

• Return of the risky asset

S(1) − S(0)
KS :=
S(0)

This is a random variable because S(1) is

a random variable.

• Return of the risk-free asset

A(1) − A(0)
KA :=
A(0)
This is a known quantity.
 • Return of a portfolio
V (1) − V (0)
KV :=
V (0)

This is also a random variable, and we con-

sider its expected value and standard devi-
ation.
 The expected return is the ex-
E(KV )
pected value of the random variable KV .
 The risk is measured by the standard
deviation
q of the return, that is
σV = Var(KV ).

Example 1) Assume A(0) = 100, A(1) = 105,

S(0) = 90 and
with prob.

100 4/5
S(1) =
90 with prob. 1/5
Let the portfolio be (x, y) = (1, 1). What are
expected return and risk?
Solution:

prob.

V (1) − V (0)  205−190 4/5
KV := 190
= 195−190
V (0) 
190 prob. 1/5

prob.

 15 4/5
= 190
 5
190 prob.
1/5

15 4 + 5 1 = 13 = 6.84%
E(KV ) = 190 5 190 5 190
 15 13 2 4  5 13 2 1
Var(KV ) = − · + − ·
190 190 5 190 190 5
16
=
1902
Var(KV ) = 190
q
σV = 4 = 2.11%

Example 2) In the setting of Example 1),

nd all portfolios (x, y) with initial capital 100
which have 3% risk.
Solution:    
 x S(1)−S(0) +y A(1)−A(0) 
Var(KV ) = Var V (0)

x2 Var
 
= V (0) 2 S(1)

Var(K V ) = 0.03 2 , V (0) = 100, and

Var S(1) = (100 − 98)2 45 + (90 − 98)2 51 = 16.

 

So, x2 = Var(KV ) V (0)2 = 16

9
Var S(1)
The portfolios are
x = 0.75, y = 1 − 0.9x = 0.325
and x = −0.75, y = 1 − 0.9x = 1.675.
Example 3) A tax-free savings account (TFSA)
allows you to invest up to $5,500 per year
such that its investment income is not taxed.
You can place the $5,500 in a bank account,
stocks or a mix of the two asset classes. Cur-
rently, most banks oer around 1% interest on
TFSA, assume that, alternatively, there is a
stock available with expected return of 5% and
risk of 20%. How would you split the $5,500
such that you get 4% expected return at min-
imal risk?
Solution:
Let z be the percentage invested in the stock.
   
x S(1) − S(0) + y A(1) − A(0)
KV =
V (0)
xS(0) yA(0)
= KS + KA
V (0)
| {z }
V (0)
| {z }
z 1−z
we nd that the expected return is given by
E(KV ) = zE(KS ) + (1 − z)KA

= 0.05z + 0.01(1 − z).

Setting this equal to 0.04 gives z = 3/4. There-
fore, the investment is 34 $5, 500 = $4,125 in
stock and 14 $5, 500 = 1, 375 in the bank ac-
count.
1.4 Options
Denitions

• A call option is a contract that gives to

its holder the right (not the obligation)
to buy one share of the stock at time 1 for
a price K xed at time 0.

• A put option is a contract that gives to

its holder the right (not the obligation)
to sell one share of the stock at time 1 for
a price K xed at time 0.

• The constant K is called the strike price

or exercise price of the option.

• We use C(t) and P (t) to denote the prices

at time t of the call option and put option,
respectively. Then
C(1) = (S(1) − K)+ = max{S(1) − K, 0},
 P (1) = (K − S(1))+ = max{K − S(1), 0}.

• An important question is determining the

prices C(0) and P (0). The calculation of
these prices is called pricing the options.

Extending the Market Model:

• We rst nd C(0). To do so, we con-

sider an extended market consisting of
the risky asset, the risk-free asset and the
call option.

• A portfolio for the extended market is a

triple (x, y, z) consisting of x units of the
risky asset, y units of the risk-free asset
and z call options. Its value is given by
V (0) = xS(0) + yA(0) + zC(0),
 V (1) = xS(1) + yA(1) + zC(1)
= xS(1) + yA(1) + z(S(1) − K)+

• The terminal price of the option C(1) can

be seen as a payo and one can think of
replicating (i.e. C(1) = aS(1) + bA(1))
it by investing in risky and risk-free as-
sets. If such a replication is possible, the
no-arbitrage principle determines the price
C(0).

Theorem 1.2 Suppose the extended market

model is arbitrage free, and that there exist
numbers a and b such that
C(1) = aS(1) + bA(1). 1.2
Then the price of the call option is
C(0) = aS(0) + bA(0). 1.3
Remarks:
• Theorem 1.2 gives a way for the pricing of
options:
(i) Find a replicating portfolio (a, b) such
that C(1) = aS(1) + bA(1) if such a
portfolio exists.
(ii) The call price is then given by the initial
value of such a portfolio:
C(0) = aS(0) + bA(0).

• Note that the theorem does not say that

there exists a replicating portfolio (a, b) such
that (1.2) holds.

• A similar result holds for put options: If

there exist a and b with
P (1) = aS(1) + bA(1),
then the put option price
P (0) = aS(0) + bA(0).
Proof: Let and b be such that (1.2) holds.
a
We assume that (1.3) does not hold and show
a contradiction.
1) First assume that C(0) < aS(0) + bA(0).
=⇒ the call option is cheaper than its replicat-
ing portfolio
=⇒ we can construct an arbitrage opportunity.
Indeed, take one call option: z = 1, x = −a and
y such that V (0) = 0, which implies

aS(0) − C(0)
0 = −aS(0)+yA(0)+C(0) =⇒ y =
A(0)

V (1) = xS(1) + yA(1) + zC(1)

  A(1)
= −aS(1) + aS(0) − C(0) + C(1)
A(0)
  A(1)
= bA(1) + aS(0) − C(0) ,
A(0)
using (1.2). The assumption C(0) < aS(0) +
bA(0) implies
  A(1)
bA(1) > C(0) − aS(0)
A(0)
and V (1) > 0 =⇒ the portfolio (x, y, z) is an
arbitrage opportunity.
2) Assume now that C(0) > aS(0) + bA(0).
=⇒ the call option is more expensive than its
replicating portfolio.
=⇒ we can construct an arbitrage opportunity.
Indeed, take one negative call option: z =
−1, x = a and y such that V (0) = 0, which
implies
C(0) − aS(0)
0 = aS(0)+yA(0)−C(0) =⇒ y =
A(0)

V (1) = xS(1) + yA(1) + zC(1)

  A(1)
= aS(1) + C(0) − aS(0) − C(1)
A(0)
  A(1)
= −bA(1) + C(0) − aS(0) ,
A(0)
using (1.2). The assumption C(0) > aS(0) +
bA(0) implies
  A(1)
−bA(1) > aS(0) − C(0)
A(0)
and V (1) > 0 =⇒ the portfolio (x, y, z) is an
arbitrage opportunity.
In conclusion, in either case C(0) < aS(0) +
bA(0) and C(0) > aS(0) + bA(0), there exist
arbitrage opportunities. Therefore, the only
arbitrage-free price is C(0) = aS(0) + bA(0).
Example 1: Call and put options on popu-
lar stocks are exchange traded. The table be-
low shows the option prices (in US dollars) on
November 4, 2016 on the Microsoft stock with
expiration date on April 21, 2017. The price
of one Microsoft share was $58.71 at the end
of November 4, 2016.
 Strike price Call price Put price
50 9.75 1.17
52.5 7.75 1.61
55 5.9 2.28
57.5 4.25 3.25
60 2.9 4.4
62.5 1.85 5.9
65 1.1 7.7
67.5 0.64 9.55
Source: Yahoo nance; check on the website
to get up-to-date values.
Note: The risk (upward and downward) of re-
turns on options is much higher than that of
returns the underlying stock.
For instance, consider call option with strike
price 60:

• if the Microsoft stock goes up 10% to 64.58

=⇒ return of option = 64.58−60−2.9
2.9 = 58%
 • if the Microsoft stock goes down 10% to
52.84
=⇒ return of option = 0−2.9
2.9 = −100%

Example 2: Let A(0) = 100, A(1) = 105,

S(0) = 50 and
with prob. p

55
S(1) =
45 with prob. 1 − p

for p ∈ (0, 1). Is it possible to replicate a call

option with strike price 50? If so, what is its
price?
Solution: From C(1) = aS(1) + bA(1), we get
5 = 55a + 105b,

0 = 45a + 105b
This implies a = 1/2 and b = −3/14. There-
fore, the call option is replicable and its price
at time t = 0 equals
1 3
C(0) = aS(0) + bA(0) = 50 − 100 ≈ 3.5714
2 14
Example 3: Consider Example 2 now for a put
option with strike price K for some K > 0. Is
it possible to replicate a put option with strike
price K ? If so, what is its price?
Solution: The equation P (1) = aS(1) + bA(1)
is equivalent to
(K − 55)+ = 55a + 105b

(K − 45)+ = 45a + 105b

To solve for a and b, we consider three dierent

cases on K .

(i) Case1: K > 55. The equations become

K − 55 = 55a + 105b

K − 45 = 45a + 105b
 which yield a = −1 and b = K/105. There-
fore, the put option is replicable with price
K 20
aS(0)+bA(0) = −50+ 100 = −50+ K
105 21

(ii) Case 2: 45 < K < 55. The equations are

0 = 55a + 105b

K − 45 = 45a + 105b
which yield a = 92 − 10
K and

−55 −33 11
b= a= + K
105 14 210
Therefore, the put option is replicable with
price
5 75
P (0) = aS(0)+bA(0) = 50a+100b = K−
21 7

(iii) Case 3: 0 6 K 6 45. The equations are

0 = 55a + 105b
 0 = 45a + 105b

which yield a = 0 and b = 0. Therefore,

the put option is replicable with price equal
to zero.

Example 4: Let A(0) = 100, A(1) = 105,

S(0) = 50 and
with prob. 0.25

55

S(1) = 50 with prob. 0.5


45 with prob. 0.25





Is it possible to replicate a call option with

strike price 50? If so, what is its price?
Solution: From C(1) = aS(1) + bA(1), we get
5 = 55a + 105b

0 = 50a + 105b

0 = 45a + 105b.
The rst two equations imply a = 1, but the
last two equations imply a = 0. Because of this
contradiction, the call option is not replicable.
What are options used for?

• Hedging:
 Shareholders can protect themselves by
buying put options
 companies can hedge against risk in for-
eign currencies
 companies can protect against non-traded
risks, for example, using weather deriva-
tives

• Speculation: risk in options is higher than

in underlying stock.
Example 1: (Speculation) Let A(0) = 100,
A(1) = 105, S(0) = 50 and
55 with prob. 4/5


S(1) =
45 with prob. 1/5

for p ∈ (0, 1). Assume there is a call option

available with a strike price 50. Suppose an
investor has 1000 initial wealth.

(i) Find the possible wealth he will have at

time t = 1 if he invests all his money on
the call option. Compute the risk.

(ii) Do the same if the investor invests all his

money on the stock.

Example 2: (Hedging) Let A(0) = 100, A(1) =

105, S(0) = 100 and
with prob.

110 1/2
S(1) =
50 with prob. 1/2
for p ∈ (0, 1). There is a put option available
with a strike price 100. Suppose an investor
owns 10 units of stocks at time t = 0.

(i) Find the possible wealth he will have on

the stock at time t = 1 if he does nothing.

(ii) Is it better if the investor buys 10 put op-

tions?

Here is some option jargon:

• Premium: The amount paid for the con-

tract initially (denoted by C(0) for call and
P (0) for put).

• Expiration date or expiry: Date on which

the option can be exercised or date on
 which the option ceases to exist or give
the holder any rights.

• Underlying asset: The nancial instru-

ment on which the option value depends.
It could be a stock, commodity, currency,
index, interest rate, etc. Its price today
is S(0) and its price at expiration date is
denoted by S(T ).

• Strike price or exercise price: The amount

at which the underlying asset can be bought
(call) or sold (put). It is denoted by K .

• In the money: An option with positive

intrinsic value. (i.e. a call option when the
asset price is above the strike, a put option
when the asset price is below the strike.)
• Out of the money: An option with no in-
trinsic value, only time value - a call option
when the asset price is below the strike, a
put option when the asset price is above
the strike.

• At the money: A call or put with a strike

that is close to the current asset level.

• Long position: A positive amount of a

quantity, or a positive exposure to a quan-
tity.

• Short position: A negative amount of a

quantity, or a negative exposure to a quan-
tity.
Bibliography

Marek Capinski and Tomasz Zastawniak, Math-

ematics for Finance, An Introduction to Finan-
cial Engineering
John C. Hull, Options, Futures, and Other
derivatives, 9th Edition, Pearson