PORTFOLIO ANALYSIS AND MANAGEMENT

SOLUTIONS TO THE PROBLEM SET FOR MODULE 3

8. Assume that an individual can either invest all of his resources in one of the two
securities A or B; or, alternatively, he can diversify his investment between the
two. The distributions of the returns are as follows:

Security A Security B

Return Probability Return Probability

________________________________________________________________________
-10 1/2 -20 1/2

50 1/2 60 1/2

Assume that the correlation between the returns from the two securities is
zero, and answer the following questions:
(a) Calculate each security's expected return, variance and standard
deviation.
(b) Calculate the probability distribution of the returns on a mixed portfolio
comprised of equal proportions of securities A and B. Also calculate the
portfolio's expected return, variance and standard deviation.
(c) Calculate the expected return and the variance of a mixed portfolio
comprised of 75% of security A and 25% of security B.

(a)
1×1 5×1
E [ R A ] =−0 . + 0. =0 . 2
2 2
2×1 6×1
E [ RB ] =−0 . +0. =0 . 2
2 2
(−0 . 1−0 . 2 )2×1 ( 0 . 5−0 . 2 )2 ×1
σ 2A = + =0 . 09
2 2
σ A =0 .3
(−0 . 2−0 . 2 )2×1 ( 0 . 6−0 .2 )2 ×1
σ 2B = + =0 . 16
2 2
σ B =0 . 4

(b) In this portfolio

w A=0 .5
w B=0 . 5

Probability distribution of the portfolio returns:

State Return Probability
1 0.5(-0.1)+0.5(-0.2)= -0.15 1/4
2 0.5(0.5)+0.5(-0.2)= 0.15 1/4
3 0.5(-0.1)+0.5(.06)= 0.25 1/4
4 0.5(0.5)+0.5(0.6)= 0.55 1/4

E [ R pf ] =w A E [ R A ]+w B E [ R B ]=0 . 5×0 . 2+0 .5×0 .2=0 .2

σ 2pf =w 2A σ 2A +w 2B σ 2B +2 w A w B ρ A , B σ A σ B=0 .5 2×0. 09+0 . 52 ×0 .16=0 . 0625
σ pf =0 . 25

c)
E [ R pf ] =w A E [ R A ]+w B E [ R B ]=0 . 75×0 . 2+0 .25×0 .2=0 .2
σ 2pf =w 2A σ 2A +w 2B σ 2B+2 w A w B ρ A , B σ A σ B=0 .75 2×0 . 09+0 . 252 ×0 .16=0 . 060625
σ pf =0 . 246

9. The securities of companies Z and Y have the following expected returns and
standard deviations:
_________________________________________________________
Company Z Company Y
__________________________________________________________
Expected Return(%) 15 35
Standard Deviation(%) 20 40
__________________________________________________________

Assume that the correlation between the returns of the two securities is 0.25.

(a) Calculate the expected return and standard deviation for the following
portfolios:
76(1) 100%Z
(2) 75%Z + 25%Y
(3) 50%Z + 50%Y
(4) 25%Z + 75%Y
(5) 100%Y

(1)
μ=μ Z =15 %
σ =σ Z=20 %

(2)
μ=wZ μ Z +wY μ Y =0. 75×15+0 .25×35=20 %
σ 2 =w 2Z σ 2Z +w 2Y σ 2Y +2×w Z wY ρ ZY σ Z σ Y
¿ 0 .75 2×20 2 +0 .25 2×402 +2×0 .75×0 . 25×0 . 25×20×40=400
σ =√ 400=20
(3)
μ=0 . 5×15+0 .5×35=25
σ 2 =0 . 52 ×202 +0 . 52 ×40 2 +2×0 . 5×0. 5×0.25×20×40=600
σ =24 . 49
(4)

μ=0 .25×15+0.75×35=30
σ 2 =0. 252 ×202 +0 . 752 ×40 2 +2×0. 25×0. 75×0 .25×20×40=1000
σ =31. 62
(5)
μ=μY =35 %
σ =σ Y =40 %

(b) Graph your results.

40
35 35 (c)
30 30
(c)
(c)
Expected Return

25 25 (c)
20 20 (c)
15 15 (c)
(c)
10
(c)
5 (c)
0 (c)
15 20 25 30 35 40 45 (c)
(c)
Std. Deviation
(c)
(c)
Which of the portfolios in part (a) is optimal? Explain.

All portfolios except for (1) are efficient. They lie on the efficient portion of the mean
variance frontier. (1) gives lower return for the same standard deviation as (2) and
therefore is not efficient.
18. The following data have been acquired for the S+P 500 and an index of Russian
stocks:
Year Market Return Russia
1998 27% 25%
1997 12% 5%
1996 -3% -5%
1995 12% 15%
1994 -3% -10%
1993 27% 30%

Does it make sense to add an index of Russian Stocks to your portfolio? Can
you reduce the risk of your portfolio - at no loss in return - by doing so?

Your decision rule should be: add Russian stocks to your portfolio if
σ SP 500
ρSP 500, RI ≺
σ RI

Using Excel, we find

σ SP500=0. 1314
σ RI =0 . 1612
ρSP 500, RI =0 .9707
σ SP500
=0. 83
σ RI

It is not beneficial to add Russian stocks to your portfolio, because their addition does not
reduce the variance of the portfolio without sacrificing the return.

22. The returns on stocks A and B are perfectly negatively correlated ( ρ AB =−1
). Stock A has an expected return of 21 % and a standard deviation of return of
40%. Stock B has a standard deviation of return of 20%. The risk-free rate of
interest is 11 %. What must be the expected return to stock B?

Since assets A and B are perfectly negatively correlated, it is possible to construct a

risk-free portfolio from them, let us call it Portfolio C. The standard deviation of C is 0:
w 2A σ 2A +w 2B σ 2B −2 w A w B σ A σ B=0
( w A σ A −w B σ B )2 =0
w A σ A−(1−w A )σ B=0
and its expected rate of return is equal to the risk-free rate. That is:
 w A E [r A ]+(1−w A )E [r B ]=r f
Combining two equations and substituting in numbers we get two equations with two
unknown variables – wA and E[rB]:
0. 4 w A−0. 2(1−w A )=0
0.21 w A +(1−w A )E [r B ]=0. 11
or
w A=0 . 2/0 .6=1/3
w B =2/3
Plug in weights into the equation for the portfolio’s C expected return:
1 2
0 .21× + ×E[ r B ]=0 .11
3 3
3
E[ r B ]=0. 04× =0 . 06
2
Therefore, expected return on asset B is 6%.

38. Consider the following two assets:

State Prob. ŘA(σ) ŘB(θ)

θ= 1 1/3 10% -5%

θ= 2 1/3 -4% -4%

θ= 3 1/3 3% 18%

What is the expectation of RA and RB? Variances? What is

~ ~
Cov( R A , R B ) ? What is ρAB?
 0 .1−0. 04 +0 . 03
E [ RA]= =0 . 03
3
−0 . 05−0 . 04+0. 18
E [ RB ] = =0 .03
3
1
√ 2 2
σ A = ( ( 0 .1−0 . 03 ) + (−0 . 04−0 . 03 ) ) =0 . 057155
3
1
√
σ B = ( (−0 . 05−0. 03 )2 + (−0 . 04−0. 03 )2 + ( 0 .18−0 . 03 )2 ) =0 .106145
3
( . 1−. 03 ) (−.05−. 03 ) + (−. 04−.03 )(−. 04−. 03 ) + ( . 03−. 03 ) ( .18−. 03 )
σ A , B= =−0. 00023
3
−0 . 00023
ρ A , B= =−0. 03846
0 . 057155×0 . 106145

39. Consider the situation of an insurance company which offers

personal injury policies to professional hockey players. The typical
payoffs to one of these policies are forecast to be the following:

State Prob. Payoff

Athlete is seriously 0.01 -5,000,000
injured
Athlete is not injured 0.99 60,000
seriously

What is the expected profit per policy? Assuming the returns

are uncorrelated policy to policy, how many policies must be
sold in order for the standard deviation of the firm's overall
portfolio of injury policies to be below $10,000?

There are two possible states for the insurance company on each policy

State 1 – athlete is seriously injured – p1 =0 .01

State 2 – athlete is not injured seriously -- p2 =0 .99

Expected payoff (or expected value of the payoff) for the each policy is

E [ payoff ] =−5 , 000 , 000×0 . 01+60 , 000×0. 99=$ 9 , 400

 All policies are identical and, therefore, the variance of the each policy is equal to
the variance of any other policy.
 All policies are uncorrelated and, therefore, the covariance between any two
policies is 0.
Recall from the class, that the variance of the equally weighted portfolio containing
securities with identical variances and covariances is

σ2 1

2
σ 2pf =
N (
+ 1− ×σ̄
N )
where σ is the variance of the each individual policy and σ̄ is the covariance
between any two policies. We know that this covariance is 0. Now, we need to find the
variance of the each individual policy:

2 2
σ 2 =(−5 , 000 , 000−9 , 400 ) ×0 .01+ ( 60 , 000−9, 400 ) ×0 . 99=253 , 476 ,000 , 000

Now, we can find the number of policies that must be sold in order to make the standard
deviation of the firm’s overall portfolio of injury policies to be below $10,000:

253 ,476 ,000 ,000

10,0002=
N
N=2,535 policies
More than 2,535 policies need to be sold to bring the standard deviation of the portfolio
below $10,000.

46. Assume N securities. The expected returns on all the securities are equal to 0.01
and the variances of their returns are all equal to 0.01. The covariances of the
returns between two securities are all equal to 0.005.
i. What are the expected return and the variance of the return on
an equally weighted portfolio of all N securities?
ii. What value will the variance approach as N gets large?
iii. What characteristic of the securities is most important when
determining the variance of a well diversified portfolio?

a. The expected return on the equally weighted portfolio is

N
∑ E [ ri ]
N ×0 . 01
E [ r ] = i=1 = =0 .01
N N

Using the formula for the variance of the equally weighted portfolio in which
variances of all the securities are equal and the covariance between any two securities
is the same, we find
 σ2 1
σ 2pf =
N ( )
+ 1− ×σ̄
N
0 . 01 1 0 . 005
σ 2pf =
N ( )
+ 1− ×0 . 005=0 . 005+
N N

N is the number of securities.

b. We need to find a limiting value for the variance of the portfolio when N becomes
very large (or approaches infinity):

lim ( 0 .005+ 0 .N005 )=0 .005

N →∞

The variance of the portfolio will approach the average covariance 0.005.

c. When determining the variance of a well-diversified portfolio, covariance

between a stock and all other stocks in the portfolio is the most important.