You are on page 1of 9

# Seminar 4 / June 3, 2015 Valuation of International Financial Assets

## Probability and Statistics

1. You are invested in two hedge funds. The probability that hedge fund Alpha gen-
erates positive returns in any given year is 60%. The probability that hedge fund
Omega generates positive returns in any given year is 70%. Assume the returns are
independent. What is the probability that both funds generate positive returns in a
given year? What is the probability that both funds lose money?
Solution: Probability that both generate positive returns = 60% 70% = 42%.
Probability that both funds lose money = (160%)(170%) = 40%30% = 12%.

2. Corporation ABC issues \$100 million of bonds. The bonds are rated BBB. The
probability that the rating on the bonds is upgraded within the year us 8%. The
probability of a downgrade is 4%. What is the probability that the rating remains
unchanged?
Solution: The sum of all three events upgrade, downgrade, and no change, must
sum to one. There is no other possible outcome. 88% + 8% + 4% = 100%.

## 3. Given the following density function:


c(100 x2 ) for 10 x 10
f (x) =
0 otherwise

## Calculate the value of c.

Solution: Given the density function, we can find c by noting that the sum of
probabilities must be equal to one:
Z Z 10  10
2 1 3
f (x)dx = c(100 x )dx = c 100x x =1
10 3 10

3
c=
4, 000
4. Given the probability density function, F (x), for 0 x 10:
x
F (x) = (20 x)
100
Check that this is a valid CDF; that is, show that F (0) = 0 and F (10) = 1. Calculate
the probability density function, f (x).
Solution:
Calculate the value of the CDF for minimum and maximum values of x:

F (0) = 0
F (10) = 1

## Sorin Dumitrescu, PhD, CFA 1

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

## Calculate the PDF by taking the first derivative of the CDF:

d 20 2x 1
f (x) = F (x) = = (10 x)
dx 100 100 50

## 5. Given the probability density function, f (x):

c
f (x) =
x
where 1 x e. Calculate the cumulative distribution function, F (x), and solve
for the constant c.
Solution: Calculate the CDF by integrating the PDF:
Z x Z x
c
F (x) = f (t)dt = dt = c[ln(t)]x1 = cln(x)
1 1 t

## We know that F (0) = 0 and F (e) = 1. F (e) = cln(e) = c 1 = c.

The CDF is expressed as:
F (x) = ln(x)

6. You own two bonds. Both bonds have a 30% probability of defaulting. Their default
probabilities are statistically independent. What is the probability that both bonds
default? What is the probability that only one bond defaults? What is the probability
that neither bond defaults?
Solution:
P[both bonds default] = 30% 30% = 9%
P[one defaults] = 2 30% (1-30%) = 42%
P[neither defaults] = (1-30%)(1-30%) = 49%

7. Your firm forecasts that there is a 50% probability that the market will be up signif-
icantly next year, a 20% probability that the market will be down significantly next
year, and a 30% probability that the market will be flat, neither up or down signif-
icantly. You are asked to evaluate the prospects of a new portfolio manager. The
manager has a long bias and is likely to perform better in an up market. Based on
past data, you believe that the probability that the manager will be up if the market
is up significantly is 80%, and that the probability that the manager will be up if
the market is down is only 10%. If the market is flat, the manager is just as likely
to be up as to be down. What is the unconditional probability that the manager is
up next year?
Solution:

## Sorin Dumitrescu, PhD, CFA 2

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

Using M to represent the market and X to represent the portfolio manager, we are
given the following information:

P [Mup ] = 50%
P [Mdn ] = 20%
P [Mf lat ] = 30%
P [Xup |Mup ] = 80%
P [Xup |Mdn ] = 10%
P [Xup |Mf lat ] = 50%

## P [Xup ] = P [Xup |Mup ] P [Mup ] + P [Xup |Mdn ] P [Mdn ]+

+ P [Xup |Mf lat ] P [Mf lat ]
P [Xup ] = 40% + 2% + 15% = 57%

8. At the start of the year, a bond portfolio consists of two bonds, each worth \$100. At
the end of the year, if a bond defaults, it will be worth \$20. If it does not default,
the bond will be worth \$100. The probability that both bonds default is 20%. The
probability that neither bond defaults is 45%. What are the mean, median, and
mode of the year-end portfolio?
Solution: We are given the probability of two outcomes:

P [V = \$40] = 20%
P [V = \$200] = 45%

At the year-end, the value of the portfolio, V, can have only one of three values,
and the sum of all probabilities must be 1. This allows us to calculate the final
probability:
P [V = \$120] = 100% 20% 45% = 35%
The mean of V is then \$140:

## = 0.20 \$40 + 0.35 \$120 + 0.45 \$200

The mode of the distribution is \$200; this is the most likely single outcome. The
median of the distribution is \$120; half of the outcomes are less than or equal to
\$120.

## 9. Given the following probability density function:

x
f (x) = s.t. 0 x 10
50
what are the mean, median, and mode of x?

## Sorin Dumitrescu, PhD, CFA 3

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

## Solution: For a continuous distribution, the mode corresponds to the maximum of

the PDF. So, the mode is equal to 10. To calculate the median, we need to find m,
such that the integral of f (x) from the lower bound of f (x), zero, to m is equal to
0.50. That is, we need to find:
Z m
x
dx = 0.50
0 50
Solve the left-hand side of the equation
Z m Z m m
m2

x 1 1 1 2 1 2
dx = xdx = x = (m 0) =
0 50 50 0 50 2 0 100 100
Set this result equal to 0.50:
m2
= 0.50
100
Which gives m = 7.07 The mean is:
Z 10 Z 10  10
x 1 2 1 1 3 1, 000
= x dx = x dx = x = = 6.67
0 50 50 0 50 3 0 150

## 10. Given the following equation:

y = (x + 5)3 + x2 + 10x
what is the expected value of y? Assume the following:
E[x] = 4
E[x2 ] = 9
E[x3 ] = 12

Solution: E[x2 ] and E[x3 ] cannot be derived from knowledge of E[x]. In this prob-
lem, E[x2 ] 6= E[x]2 and E[x3 ] 6= E[x]3 . Find the expected value of y:
E[y] = E[(x + 5)3 + x2 + 10x] = E[x3 + 16x2 + 85x + 125]
E[y] = E[x3 ] + E[16x2 ] + E[85x] + E[125]
= E[x3 ] + 16E[x2 ] + 85E[x] + 125

## Which gives us:

E[y] = 12 + 16 9 + 85 4 + 125 = 621
11. Assume that a random variable Y has a mean of zero and a standard deviation of
one. Given two constants, and , calculate the expected value of X1 and X2 , where
X1 and X2 are defined as:
X1 = Y +
X2 = (Y + )

## Sorin Dumitrescu, PhD, CFA 4

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

Solution:
The expected value of X1 is :

## E[X2 ] = E[(Y + )] = E[Y + ]

= E[Y ] + = 0 + =

Multiplying a standard normal variable by a constant and then adding another con-
stant produces a different result than if we first add and then multiply.

12. X is a random variable. X has an equal probability of being -1, 0, or +1. What is
the correlation between X and Y if Y = X 2 .
Solution:
We have:
1
P [X = 1] = P [X = 0] = P [X = 1] =
3
Y = X2

## First, we calculate the mean of both variables:

1 1 1
E[X] = (1) + (0) + (1) = 0
3 3 3
1 1 1 1 1 2
E[Y ] = (12 ) + (02 ) + (12 ) = (1) + (1) =
3 3 3 3 3 3
The covariance can be found as:

## Cov[X, Y ] = E[(X E[X])(Y E[Y ])]

   
1 2 1 2
Cov[X, Y ] = (1 0) 1 + (0 0) 0
3 3 3 3
 
1 2
+ (1 0) 1 =0
3 3
Because the covariance is 0, the correlation is also zero. There is no need to calculate
the variances and standard deviations.

## 13. Prove the following result:

E[(X )3 ] = E[X 3 ] 3 2 3

Solution:

## Sorin Dumitrescu, PhD, CFA 5

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

## We start by multiplying out the terms inside the expectation:

E[(X )3 ] = E[X 3 3X 2 + 32 X 3 ]

Next, we separate the terms inside the expectation operator and move any constants,
namely , outside the operator: 1

## E[X 3 3X 2 + 32 X 3 ] = E[X 3 ] 3E[X 2 ] + 32 E[X] 3

E[X] is simply the mean, . For E[X 2 ], we reorganize our equation for variance, as
follows:

2 = E[X 2 ] 2
E[X 2 ] = 2 + 2

Substituting these results into our equation and collecting terms, we arrive at the
final equation:

E[(X )3 ] = E[X 3 ] 3( 2 + 2 ) + 32 3
= E[X 3 ] 3 2 3

14. Given two random variables, XA and XB , with corresponding means A and B and
standard deviations A and B , prove that the variance of XA plus XB is:

V ar[XA + XB ] = A2 + B2 + 2AB A B

## where AB is the correlation between XA and XB .

Solution:
First we note that the expected value of XA plus XB is just the sum of the means.

## V ar[XA + XB ] = E[(XA + XB E[XA + XB ])2 ]

= E[((XA A ) + (XB B ))2 ]

## V ar[XA + XB ] = E[(XA A )2 + (XB B )2 + 2(XA A )(XB B )]

= E[(XA A )2 ] + E[(XB B )2 ] + 2E[(XA A )(XB B )]
= A2 + B2 + 2Cov[XA , XB ]
= A2 + B2 + 2AB A B
1
Pn n

(x + y)n = k=0 k xnk y k

## Sorin Dumitrescu, PhD, CFA 6

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

15. Assume we have four bonds, each with a 10% probability of defaulting over the next
year. The event of default for any given bond is independent of the other bonds
defaulting. What is the probability that zero, one, two, three, or all of the bonds
default? What is the mean number of defaults? The standard deviation?
Solution:
We can calculate the probability of each possible outcome as follows:

n

# of defaults2 k
pk (1 p)k Probability

0 1 65.61% 61.61%
1 4 7.29% 29.16%
2 6 0.81% 4.86%
3 4 0.09% 0.36%
4 1 0.01% 0.01%
100.00%

We can calculate the mean number of default two ways. The first is to use the formula
for the mean:
= np = 4 10% = 0.40
On average there are 0.40 defaults. The other way we could arrive at this result is
to use the probabilities from the table:
4
X
= pi xi = 0.40
i=0

Solution (cont.): To calculate the standard deviation, we also have two choices:

## 2 = np(1 p) = 4 10%(1 10%) = 0.36

= 0.60

As with the mean, we could also use the probabilities from the table:
4
X
2
= pi (xi )2 = 0.60
i=0

16. For a uniform distribution with a lower bound x1 and an upper bound x2 , prove that
the formulas for calculating the mean and variance are:
1
= (x2 + x1 )
2
1
2 = (x2 x1 )2
12

## Sorin Dumitrescu, PhD, CFA 7

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

Solution:
For the mean:
Z x2 Z x2  x 2
1 1
= cxdx = c xdx = c x2 = c (x22 x21 )
x1 x1 2 x1 2

## We know that c = 1/(x2 x1 )3 , therefore:

1
= (x2 + x1 )
2
For the variance:
Z x2 Z x2  x2
2 2 1
= c(x ) dx = c (x 2x + )dx = c x3 x2 + 2 x
2 2
x1 x1 3 x1

## Substituting for c and we get:

1
2 = (x2 x1 )2
12

17. Prove that the normal distribution is a proper probability distribution. That is, show
that: Z
1 (x)2
e 22 dx = 1
2 2
Use: Z
2
ex dx =

## Solution: Using integration by substitution, define a new variable y and solve:

x
y=
2

dx = 2dy
Z
1 (x)2 1 2 1
e 22 dx = ey dy = =1
2

18. Prove that the mean of the normal distribution is . That is, show that:
Z
1 (x)2
x e 22 dx =
2 2
3
R x2
x1
cdx = [cx]xx21 = c(x2 x1 ) = 1

## Sorin Dumitrescu, PhD, CFA 8

Seminar 4 / June 3, 2015 Valuation of International Financial Assets

## Solution: Using the same substitution as the previous question:

x
y=
2

dx = 2dy
Z Z
1
(x) 2
1 2
x e 2 2
dx = ( 2y + )ey dy
2
Z
2 y2
Z
2
= ye dy + ey dy

2 1 2
= [ yey ] + =
2

19. Prove that the variance of a normal distribution is 2 . You may find the following
result useful: Z
2 1
x2 ex dx =
2

Solution:
x
y=
2

dx = 2dy
Z
2 2 2 y2
Z
2 1
(x)2
V ar[x] = (x ) e 2 dx =
2
y e dy
2

We know that:
1
Z
2
x2 ex dx =
2
Using this result:
2 2 1
V ar[x] = = 2
2