Managing risk and Uncertainty (EPM944)– Solutions to Examples 2

Question 1

Consider a random variable X with probability distribution as follows:

P (X = −5) = 0.3, P (X = 0) = 0.2, P (X = 5) = 0.3, P (X = 8) = 0.2.

Calculate VaR0.75 (X) and CVaR0.75 (X).

Solution: Write the cdf function as

FX (t) = 0 for t ∈ (∞, −5)

= 0.3 for t ∈ [−5, 0)
= 0.5 for t ∈ [0, 5)
= 0.8 for t ∈ [5, 8)
= 1.0 for t ∈ [8, ∞)

Hence:
VaR0.75 = inf{t : FX (t) ≥ 0.75} = 5

and
5 × 0.3 + 8 × 0.2 3.1
CVaR0.75 = E[X|X ≥ 5] = = = 6.2
0.5 0.5

Question 2

Consider an investment where with probability 0.9 you gain 10% profit of your total investment and with
probability 0.1 you lose 50% of your total investment. Suppose you have a total asset of £20, 000 for investment.
Calculate VaR0.9 . Suppose now there is another investment opportunity which has the same probability of gain
and loss. Calculate VaR0.9 of the combined investment if the fund is evenly split in the investments.

Solution: Let L represent the loss r.v. Then the cdf of L is:

FL (t) = 0 for t ∈ (−∞, −2, 000]

= 0.9 for t ∈ (−2, 000, 10, 000]
= 1.0 for t ∈ (10, 000, ∞)

Hence
VaR0.9 = inf{t : FL (t) ≥ 0.9} = −2, 000

When the fund is split equally into two investments we have the following cases:

• Neither investment fails: L = −1, 000 − 1, 000 = −2, 000 with probability 0.9 × 0.9 = 0.81.

• One investment fails, the other does not: L = 5, 000 − 1, 000 = 4, 000 with probability 2 × 0.9.1 = 0.18.

• Both investments fail: L = 5, 000 + 5, 000 = 10, 000 with probability 0.1 × 0.1 = 0.01. Hence:

FL (t) = 0 for t ∈ (−∞, −2, 000]

= 0.81 for t ∈ (−2, 000, 4, 000]
= 0.99 for t ∈ (4, 000, 10, 000]
= 1.0 for t ∈ (10, 000, ∞)

1
and
VaR0.9 = inf{t : FL (t) ≥ 0.9} = 4, 000

Question 3

Explain the following:

(a) A utility function is increasing.

(b) A beggar will value £1 more than a millionaire; a loss of £10, 000 is more significant for a small than for
a large company.

Solution:

(a) (The basic attribute of a utility function is that it an increasing function as normally everyone values more
money over less money, so u0 (x) > 0. The fact that u0 (x) 6= 0 means that there is non-satiation, i.e. the
agent never becomes completely satisfied and will always prefer more to less.

(b) In the case of the beggar and the millionaire valuation of £1, we can see that as wealth increases, each
additional £1 has a lower perceived value. This is not surprising and is known as decreasing marginal
utility, that is u00 (x) < 0.

Question 4

The common utility functions are:

(a) Exponential: u(x) = −e−ax or u(x) = 1 − e−ax , a > 0.

(b) Logarithmic: u(x) = ln(1 + x).

xα −1
(c) Power: u(x) = γ ,0 < α < 1.

x1−ρ
(d) Iso-elastic: 1−ρ , 0 < ρ < 1.

Show that all these functions are concave. Can you construct a strictly increasing convex utility function?

Solution: (a) u(x) = −e−ax . Then u0 (x) = ae−ax > 0 and u00 (x) = −a2 e−ax < 0 so function is increasing
and concave. (b) u(x) = ln(1 + x) (x > −1). Then u0 (x) = 1+x
1
> 0 and u00 (x) = − (1+x)
1
2 < 0 so function
α
x −1 0
is increasing and concave. (c) u(x) = γ , 0 < α < 1, γ > 0 (x > 0). Then u (x) = αγ xα−1 > 0 and
α(α−1) α−2 x1−ρ
u00 (x) = γ x < 0. Increasing concave function for 0 < α < 1 (convex for α > 1). (d) 1−ρ , 0<ρ<1
0 x−ρ 00 ρ
(x > 0): u (x) = 1−ρ > 0, u (x) = − 1−ρ x−1−ρ < 0. Increasing concave function.
ax α
Functions u(x) = e (α > 0) and u(x) = x (α > 1) are strictly increasing and convex.

Question 5

Consider cumulative distribution functions of random variables X and Y :

FX (x) = 0, if x < 2.5 FY (x) = 0, if x < 2

= 0.4, if 2.5 ≤ x < 3.5 = 0.5, if 2 ≤ x < 3
= 1, if 3.5 ≤ x = 1, if 3 ≤ x

2
Show that: (a) FX dominates FY in first order. (b) For arbitrary utility function u, E[u(X)] ≥ E[u(Y )]. Assume
u(0) = 0. (c) If Fx is changed to:

FX (x) = 0, if x < 2.5

= 0.6, if 2.5 ≤ x < 3.5
= 1, if 3.5 ≤ x

discuss first order dominance and second order dominance conditions. Show that E[u(X)] ≥ E[u(Y )] holds for
u(x) = ln(1 + x). Does the inequality hold for u(x) = x or u(x) = x2 ?

Solution: (a) It is easy to see that for every x ∈ R: FX (x) ≤ FY (x). Equivalently P {X ≤ x} ≤ P {Y ≤ x} or
P {X > x} > P {Y > x}, and hence X dominates Y in first-order. (b) Every utility function is non-decreasing.
By EUT (see notes) E[U (X)] ≥ E[U (Y )] if and only if X dominates Y in first order.

1.0

FY(x) FX(x)

0.5
0.4

2.0 2.5 3.0 3.5 x

Rx Rx
Clearly X does not dominate Y first order. The plots of IX (x) = −∞ FX (x)dx and IY (x) = −∞ FY (x)dx are
shown in the Figure below. Clealy X dominates Y in second order.

IY(x) IX(x)
1
1

0.6
0.5

2.0 2.5 3.0 3.5 x

Since u(x) = ln(1 + x) is a concave increasing function by EUT (see lecture notes) E[u(X)] ≥ E[u(Y )]. The
inequality may not hold for u(x) = x2 since this function is convex.

Question 6
A risk averse agent, whose utility is given by U (x) = ln x and whose wealth is £50, 000 is faced with a potential
loss of £10, 000 with a probability of 0.1. What is the maximum premium they would be willing to pay to
protect themselves against this loss? What is the minimum premium that an insurer, with the same utility
function and with wealth £1, 000, 000 will be willing to charge to cover this loss?

Solution: The agent’s expected wealth without insurance is:

E[WL ] = 50, 000 × 0.9 + 40, 000 × 0.1 = 49, 000

3
while their expected utility of wealth without insurance is:

W[U (WL )] = ln(50, 000) × 0.9 + ln(40, 000) × 0.1 = 10.797

To identify the maximum premium we need to equate:

ln(50, 000 − pmax ) = 10.797 ⇒ pmax = 1, 103

As for the insurer:

ln(1, 000, 000) = 0.9 × ln(1, 000, 000 + pmin ) + 0.1 × ln(1, 000, 000 + pmin − 10, 000)

from which the value can be found numerically as pmin ≈ 1046.

Question 7

Consider the St. Petersburg paradox. Suppose that most people will be willing to pay no more than 10 dollars
for playing the game. Discuss what kind of utility function people may have. Suppose that the utility function
takes the following form:
1
u(x) = a −
1+x
Work out the value of a.

Solution: Here
  
1 1 1 1 1 1
E[u(X)] = × u(1) + × u(2) + × u(4) + (. . .)
2 2 2 2 2 2
1 1 1 1
= × u(1) + × u(2) + × u(4) + × u(8) + . . .
2 4 8 16
Thus (approximately)
       
1 1 1 1 1 1 1 1
a− + a− + a− + a− = 10
2 1+1 4 1+2 8 1+4 16 1+8
or
15a
− 0.365 = 10 ⇒ a ≈ 10.365
16

Question 8

Consider a portfolio with two assets with return rates denoted by ξ1 and ξ2 which satisfy:

ξ1 /ξ2 0.9 1.1 Marginal prob.

0.9 0.1 0.5 0.6
1.1 0.3 0.1 0.4
Marginal prob. 0.4 0.6

Formulate a portfolio optimization problem in which the cost function to be maximized is the expected return
minus a constant multiple of the variance of the portfolio return (the constant which multiplies the variance is
known as the risk parameter). Assume that the portfolio weights add up to one and no short-selling is allowed.
Discuss how the risk parameter affects the optimal investment policy and expected return in the Markowitz
model as it decreases to zero.

Solution: The probability joint mass function of (ξ1 , ξ2 ) is: (0.9, 0.9) with probability 0.1, (0.9, 1.1) with
probability 0.5, (1.1, 0.9) with probability 0.3 and (1.1, 1.1) with probability 0.1. Hence:

E[ξ1 ] = 0.9 × 0.6 + 1.1 × 0.4 = 0.98, E[ξ2 ] = 0.9 × 0.4 + 1.1 × 0.6 = 1.02

4
Also

E[ξ1 ξ2 ] = 0.9 × 0.9 × 0.1 + 0.9 × 1.1 × 0.5 + 1.1 × 0.9 × 0.3 + 1.1 × 1.1 × 0.1 = 0.9940
Var(ξ1 ) = 0.92 × 0.6 + 1.12 × 0.4 − 0.982 = 0.0096
Var(ξ2 ) = 0.92 × 0.4 + 1.12 × 0.6 − 1.022 = 0.0096
Cov(ξ1 , ξ2 ) = 0.9940 − 0.98 × 1.02 = −0.0056

Thus Markowitz model with risk parameter η corresponds to the QP optimization problem:
! !
  0.0096 −0.0056 w1
max 0.98w1 + 1.02w2 − η w1 w2 : w1 + w2 = 1, w1 ≥ 0, w2 ≥ 0
−0.0056 0.0096 w2

When η is driven to zero we get Linear Programming problem: max 0.98w1 + 1.02w2 s.t. w1 + w2 = 1, w1 ≥ 0,
w2 ≥ 0 whose optimal solution is clearly w1∗ = 0, w2∗ = 1.

Question 9

Consider the function F (x, ξ) in which x is a decision vector and ξ a random variable. Give examples which shows
that maxx F (x, E[ξ]) ≥ maxx E[F (x, ξ)] when F (x, ξ) is concave in ξ and minx F (x, E[ξ]) ≤ minx E[F (x, ξ)] when
F (x, ξ) is convex in ξ.

Solution: Consider F (x, ξ) = −(x − ξ)2 where ξ is uniformly distributed in [−1, 1] and x ∈ [0, 1]. Clearly F is
concave in ξ. Then
F (x, E[ξ]) = −(x − E[ξ])2 = −x2

and hence
max F (x, E[ξ]) = max (−x2 ) = 0
x∈[0,1] x∈[0,1]

Also

max E[F (x, ξ)] = max E[−(x − ξ)2 ] = max E[−x2 − ξ 2 + 2xξ]
x∈[0,1] x∈[0,1] x∈[0,1]
 
1 1
= max −x2 − E[ξ 2 ] + 2xE[ξ] = max −x2 −


=−
x∈[0,1] x∈[0,1] 3 3

and thus
1
max F (x, E[ξ]) = 0 > max E[F (x, ξ)] = −
x∈[0,1] x∈[0,1] 3
2
For a convex example we can use F (x, ξ) = (x − ξ) .

5
 Table of Normal Distribution cdf Φ(z)
Z z
1 2
Φ(z) = √ e−ξ /2 dξ ; Φ(−z) = Φ(z)
2π −∞

z 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.0 0.5000 0.5040 0.5080 0.5120 0.5160 0.5199 0.5239 0.5279 0.5319 0.5359
0.1 0.5398 0.5438 0.5478 0.5517 0.5557 0.5596 0.5636 0.5675 0.5714 0.5753
0.2 0.5793 0.5832 0.5871 0.5910 0.5948 0.5987 0.6026 0.6064 0.6103 0.6141
0.3 0.6179 0.6217 0.6255 0.6293 0.6331 0.6368 0.6406 0.6443 0.6480 0.6517
0.4 0.6554 0.6591 0.6628 0.6664 0.6700 0.6736 0.6772 0.6808 0.6844 0.6879
0.5 0.6915 0.6950 0.6985 0.7019 0.7054 0.7088 0.7123 0.7157 0.7190 0.7224
0.6 0.7257 0.7291 0.7324 0.7357 0.7389 0.7422 0.7454 0.7486 0.7517 0.7549
0.7 0.7580 0.7611 0.7642 0.7673 0.7704 0.7734 0.7764 0.7794 0.7823 0.7852
0.8 0.7881 0.7910 0.7939 0.7967 0.7995 0.8023 0.8051 0.8078 0.8106 0.8133
0.9 0.8159 0.8186 0.8212 0.8238 0.8264 0.8289 0.8315 0.8340 0.8365 0.8389
1.0 0.8413 0.8438 0.8461 0.8485 0.8508 0.8531 0.8554 0.8577 0.8599 0.8621
1.1 0.8643 0.8665 0.8686 0.8708 0.8729 0.8749 0.8770 0.8790 0.8810 0.8830
1.2 0.8849 0.8869 0.8888 0.8907 0.8925 0.8944 0.8962 0.8980 0.8997 0.9015
1.3 0.9032 0.9049 0.9066 0.9082 0.9099 0.9115 0.9131 0.9147 0.9162 0.9177
1.4 0.9192 0.9207 0.9222 0.9236 0.9251 0.9265 0.9279 0.9292 0.9306 0.9319
1.5 0.9332 0.9345 0.9357 0.9370 0.9382 0.9394 0.9406 0.9418 0.9429 0.9441
1.6 0.9452 0.9463 0.9474 0.9484 0.9495 0.9505 0.9515 0.9525 0.9535 0.9545
1.7 0.9554 0.9564 0.9573 0.9582 0.9591 0.9599 0.9608 0.9616 0.9625 0.9633
1.8 0.9641 0.9649 0.9656 0.9664 0.9671 0.9678 0.9686 0.9693 0.9699 0.9706
1.9 0.9713 0.9719 0.9726 0.9732 0.9738 0.9744 0.9750 0.9756 0.9761 0.9767
2.0 0.9772 0.9778 0.9783 0.9788 0.9793 0.9798 0.9803 0.9808 0.9812 0.9817
2.1 0.9821 0.9826 0.9830 0.9834 0.9838 0.9842 0.9846 0.9850 0.9854 0.9857
2.2 0.9861 0.9864 0.9868 0.9871 0.9875 0.9878 0.9881 0.9884 0.9887 0.9890
2.3 0.9893 0.9896 0.9898 0.9901 0.9904 0.9906 0.9909 0.9911 0.9913 0.9916
2.4 0.9918 0.9920 0.9922 0.9925 0.9927 0.9929 0.9931 0.9932 0.9934 0.9936
2.5 0.9938 0.9940 0.9941 0.9943 0.9945 0.9946 0.9948 0.9949 0.9951 0.9952
2.6 0.9953 0.9955 0.9956 0.9957 0.9959 0.9960 0.9961 0.9962 0.9963 0.9964
2.7 0.9965 0.9966 0.9967 0.9968 0.9969 0.9970 0.9971 0.9972 0.9973 0.9974
2.8 0.9974 0.9975 0.9976 0.9977 0.9977 0.9978 0.9979 0.9979 0.9980 0.9981
2.9 0.9981 0.9982 0.9982 0.9983 0.9984 0.9984 0.9985 0.9985 0.9986 0.9986
3.0 0.9987 0.9987 0.9987 0.9988 0.9988 0.9989 0.9989 0.9989 0.9990 0.9990
3.1 0.9990 0.9991 0.9991 0.9991 0.9992 0.9992 0.9992 0.9992 0.9993 0.9993
3.2 0.9993 0.9993 0.9994 0.9994 0.9994 0.9994 0.9994 0.9995 0.9995 0.9995
3.3 0.9995 0.9995 0.9995 0.9996 0.9996 0.9996 0.9996 0.9996 0.9996 0.9997
3.4 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9998
3.5 0.9998 0.9998 0.9998 0.9998 0.9998 0.9998 0.9998 0.9998 0.9998 0.9998
3.6 0.9998 0.9998 0.9999 0.9999 0.9999 0.9999 0.9999 0.9999 0.9999 0.9999