MATH5320: Discrete Time Finance

Lecture 7

G. Aivaliotis, c University of Leeds

March 2021

1 / 23
 In the last lecture

In this lecture
• Introduction to Optimal Investment: Video 2
• Utility Functions: Video 2
• Optimal portfolios and arbitrage: Video 3
• The 2-step approach: Videos 4 and 5

2 / 23
 General single period market model

Ω := {ω1 , ..., ωk }, P

Assumption: P(ω) > 0 for all ω ∈ Ω.

Asset prices:

S1i : Ω → R, i = 1, . . . , n.

Money market account:

B1 = (1 + r), B0 = 1.

3 / 23
 Utility function

A continuously differentiable function u : R+ → R is called a risk

averse utility function if it has the following three properties:
1 limx→0 u0 (x) = +∞ and limx→∞ u0 (x) = 0
2 u0 (x) > 0 for all x ∈ R
3 u is strictly concave in the way that
u(λx + (1 − λ)y) > λu(x) + (1 − λ)u(y).

Measurement of the performance: (no risk neutral measure here!!!)

Pk
E(u(V1 (x, φ))) = i=1 P(ωi )u(V1 (x, φ)(ωi )).

4 / 23
 Arrow-Pratt Coefficients of Risk Aversion

00
Coefficient of Absolute Risk Aversion: a(x) = − UU 0 (x)
(x)

Absolute risk-aversion measures the actual amount in pounds that

an investor will choose to hold in risky assets, given a certain
wealth level x.

00
Coefficient of Relative Risk Aversion: r(x) = − xU (x)
U 0 (x)

The coefficient of relative risk-aversion measures the percentage of

wealth held in risky assets, for a given wealth level x.

5 / 23
 Examples

• Logarithmic utility: u(x) = ln(x)

• Power utility: u(x) = γ1 xγ , γ ∈ (0, 1).

6 / 23
 Optimal portfolio problem

For a given initial investment
x find a trading strategy φ

that maximizes E u V1 (x, φ) .

A trading strategy (x, φ∗ ) is called a solution for the optimal

portfolio problem with the initial investment x and utility function
u if
E(u(V1 (x, φ∗ )) = max E u(V1 (x, φ))


φ

7 / 23
 Optimal strategy vs. arbitrage

Theorem 1. If there exists a solution to the optimal portfolio

problem then there can not exist an arbitrage in the market.

8 / 23
 Optimal strategy vs. arbitrage

Theorem 2. Let (x, φ∗ ) be a solution to the optimal portfolio

problem with initial wealth x and utility function u, then the
measure Q defined by

P(ω)u0 (V1 (x, φ∗ )(ω))

Q(ω) = , ω ∈ Ω,
E(u0 (V1 (x, φ∗ ))

is a risk neutral probability measure.

9 / 23
 Example

Suppose that n = 2, k = 3, r = 91 , Ŝ01 = 6, Ŝ02 = 10 and that the

discounted price process is given by the following table:
ω1 ω2 ω3
Ŝ11 6 8 4
Ŝ12 13 9 8
Take the exponential utility function with λ = 1, i.e.

u(x) = 1 − exp(−x).

Find an optimal portfolio φ.

10 / 23
 Two-step approach

Step 1:
Compute the maximizer V1 of the function

V 7→ Eu(V ),

where V is allowed to take values in a certain feasible set.

Step 2:
Compute a trading strategy which has the maximizer V1 com-
puted in step 1 as value at time t = 1.

11 / 23
 Complete market model

• one risk neutral measure Q,

• all contingent claims are attainable,
• the replication price of a contingent claim X
 X 
EQ .
1+r
• the set of wealths attainable from initial investment x > 0:

n   o
1
Wx := W ∈ Rk |EQ 1+r W =x .

12 / 23
 Optimisation problem

maximize E(u(W ))
subject to W ∈ Wx .

maximize E(u(W ))
W
subject to EQ = x.
1+r

maximize E(u(W ))
W
subject to EQ − x = 0.
1+r

13 / 23
 The Lagrange function

maximize E(u(W ))
W
subject to EQ − x = 0.
1+r

   
1
L(W, λ) := E(u(W )) − λ EQ 1+r W −x .

Here W is a vector in Rk , and λ is a real number.

14 / 23
 Solving...

1
   
L(W, λ) := E(u(W )) − λ EQ W −x
1+r

Rewrite:
k
1
X   
L(W, λ) = P(ωi ) u(W (ωi )) − λ L(ωi ) W (ωi ) − x ,
i=1
1+r

where
Q(ω)
L(ω) = .
P(ω)

15 / 23
 Inverse of u0

Let us denote the inverse function of u0 (x) with I(x), i.e.

u0 (I(x)) = x = I(u0 (x)).

Properties:
1 I is strictly decreasing,
2 limx→0 I(x) = ∞,
3 limx→∞ I(x) = 0.

16 / 23
 Final result

To find λ solve (the solution is unique and always exists):

  
1 L
EQ 1+r I λ 1+r = x.

Define
 
W (ω) = I λ L(ω)
1+r .

Move to step two: find a trading strategy that replicates W .

17 / 23
 Example

We use the utility function u(x) = ln(x). In this case u0 (x) = 1

x and
therefore I(x) = x1 as well.
This is λ:
1
λ= .
x

This is W :
1+r
W =x .
L

Move to step two: find a trading strategy that replicates W .

 
1+r
Maximal happiness: E(u(W )) = ln(x) + E ln .
L

18 / 23
 Incomplete market model

• more than one risk neutral probability measure,

• not every contingent claim is attainable:

Thecontingent
 claim X is attainable if and only if
1
EQ 1+r X takes the same value for all Q ∈ M.

• the set of wealths attainable from the initial investment x > 0:

n   o
1
Wx := W ∈ Rk |EQ 1+r W = x for all Q ∈ M .

19 / 23
 Crucial theorem
Theorem. There exist finitely many probability measures Qi ,
i = 1, ..., l (not necessarily in M) such that the space Wx is given
by

n   o
1
Wx := W ∈ Rk |EQi 1+r W = x for all i = 1, ..., l .

maximize E(u(W ))
W
subject to E Q1 − x = 0,
1+r
..
.
W
E Ql − x = 0.
1+r
20 / 23
 Lagrange function

maximize E(u(W ))
W
 
subject to EQi = x, i = 1, ..., l.
1+r

l
Li W
X    
L(W, λ) = E(u(W )) − λi E −x
i=1
1+r
with
Qi (ω)
Li (ω) = and λ = (λ1 , ..., λl )T .
P(ω)

21 / 23
 Final result

To find λ solve:

  
λ1 L1 +...+λl Ll
E Li I 1+r = (1 + r)x, i = 1, . . . , l.

Define
P 
l Li (ω)
W (ω) = I i=1 λi 1+r .

Move to step two: find a trading strategy that replicates W .

22 / 23