CHAPTER 7

AN INTRODUCTION TO PORTFOLIO MANAGEMENT

Answers to Questions

1. Investors hold diversified portfolios in order to reduce risk, that is, to lower the variance
of the portfolio, which is considered a measure of risk of the portfolio. A diversified
portfolio should accomplish this because the returns for the alternative assets should not
be correlated so the variance of the total portfolio will be reduced.

2. The covariance is equal to ΣE[(Ri - E(Ri))(Rj - E(Rj))] and shows the absolute amount of
comovement between two series. If they constantly move in the same direction, it will be
a large positive value and vice versa. Covariance is important in portfolio theory because
the variance of a portfolio is a combination of individual variances and the covariances
among all assets in the portfolio. It is also shown that in a portfolio with a large number
of securities the variance of the portfolio becomes the average of all the covariances.

3. Similar assets like common stock or stock for companies in the same industry (e.g., auto
industry) will have high positive covariances because the sales and profits for the firms
are affected by common factors since their customers and suppliers are the same.
Because their profits and risk factors move together you should expect the stock returns
to also move together and have high covariance. The returns from different assets will not
have as much covariance because the returns will not be as correlated. This is even more
so for investments in different countries where the returns and risk factors are very
unique.

4. The covariance between the returns of assets i and j is affected by the variability of these
two returns. Therefore, it is difficult to interpret the covariance figures without taking
into account the variability of each return series. In contrast, the correlation coefficient is
obtained by standardizing the covariance for the individual variability of the two return
series, that is: rij = covij/(σiσj)
Thus, the correlation coefficient can only vary in the range of -1 to +1. A value of +1
would indicate a perfect linear positive relationship between Ri and Rj.

5. The efficient frontier has a curvilinear shape because if the set of possible portfolios of
assets is not perfectly correlated the set of relations will not be a straight line, but is
curved depending on the correlation. The lower the correlation the more curved.
6. Expected Rate B
Of Return C F

A
D
E

Expected Risk (σ of Return)

7-1
 A portfolio dominates another portfolio if: 1) it has a higher expected return than another
portfolio with the same level of risk, 2) a lower level of expected risk than another
portfolio with equal expected return, or 3) a higher expected return and lower expected
risk than another portfolio. For example, portfolio B dominates D by the first criterion. A
dominates D by the second, and C dominates D by the third.

The Markowitz efficient frontier is simply a set of portfolios that is not dominated by any
other portfolio, namely those lying along the segment E-F.

7. The necessary information for the program would be:

1) the expected rate of return of each asset
2) the expected variance of return of each asset
3) the expected covariance of return of all pairs of assets under consideration.

8. Investors’ utility curves are important because they indicate the desired tradeoff by
investors between risk and return. Given the efficient frontier, they indicate which
portfolio is preferable for the given investor. Notably, because utility curves differ one
should expect different investors to select different portfolios on the efficient frontier.

9. The optimal portfolio for a given investor is the point of tangency between his set of
utility curves and the efficient frontier. This will most likely be a diversified portfolio
because almost all the portfolios on the frontier are diversified except for the two end
points - the minimum variance portfolio and the maximum return portfolio. These two
could be significant.

10. The utility curves for an individual specify the trade-offs she is willing to make between
expected return and risk. These utility curves are used in conjunction with the efficient
frontier to determine which particular efficient portfolio is the best for a particular
investor. Two investors will not choose the same portfolio from the efficient set unless
their utility curves are identical.

11. The hypothetical graph of an efficient frontier of U.S. common stocks will have a curved
shape (see the graph in the answer to question 6, above). Adding U.S. bonds to the
portfolio will likely generate a new efficient frontier that is shifted up (or to the left) of
the original stock-only frontier. The reason for the shift is that we expect bonds to be less
correlated with stocks, thereby creating additional diversification potential.
The third frontier (which includes international securities) will likely display another shift
upward or to the left for similar reasons—lower correlation potential resulting in
additional diversification potential.

12. The portfolio constructed containing stocks L and M would have the lowest standard
deviation. As demonstrated in the chapter, combining assets with equal risk and return
but with low positive or negative correlations will reduce the risk level of the portfolio.

13. Answer is C as expected return = (.15)(.50) + (.10).40) + (.06)(.10) = 12.1%

7-2
14. Answer is A as adding an investment that has a correlation of -1.0 will achieve maximum
risk diversification.

7-3
 CHAPTER 7

Answers to Problems

1. [E(Ri)] for Lauren Labs

Possible Expected
Probability Returns Return
0.10 -0.20 -0.0200
0.15 -0.05 -0.0075
0.20 0.10 0.0200
0.25 0.15 0.0375
0.20 0.20 0.0400
0.10 0.40 0.0400
E(Ri) = 0.1100

2.
Security Portfolio
Market Return Return
Stock Value Weight (Ri) W i x Ri
Disney $15,000 0.160 0.14 0.022
Starbucks 17,000 0.181 -0.04 -0.007
Harley Davidson 32,000 0.340 0.18 0.061
Intel 23,000 0.245 0.16 0.039
Walgreens 7,000 0.074 0.05 0.004
TOTAL 94,000 1.0000 0.119

3. Madison Sophie [Ri-E(Ri)] x

Month Cookies(Ri) Electric(Rj) Ri-E(Ri) Rj-E(Rj) [Rj-E(Rj)]
1 -.04 .07 -.057 .06 -.0034
2 .06 -.02 .043 -.03 -.0013
3 -.07 -.10 -.087 -.11 .0096
4 .12 .15 .103 .14 .0144
5 -.02 -.06 -.037 -.07 .0026
6 .05 .02 .033 .01 .0003
Sum .10 .06 .0222

3(a). E(RMadison) = .10/6 = .0167 E(RSophie) = .06/6 = .01

3(b).
σ Madison = .0257 / 5 = .0051 = .0717

σ Sophie = .04120 / 5 = .0082 = .0908

7-4
3(c). COVij = 1/5 (.0222) = .0044

3(d).
.0044
rij =
(.0717) (.0908)
.0044
=
.006510
= .6758

One should have expected a positive correlation between the two stocks, since they tend
to move in the same direction(s). Risk can be reduced by combining assets that have low
positive or negative correlations, which is not the case for Madison Cookies and Sophie
Electric.

4. E(R1) = .15 E(σ1) = .10 w1 = .5

E(R2) = .20 E(σ2) = .20 w2 = .5
E(Rport) = .5(.15) + .5(.20) = .175

If r1,2 = .40
σ p = (.5) 2 (.10) 2 + (.5) 2 (.20) 2 + 2(.5)(.5)(.10)(.20)(.40)

= .0025 + .01 + .004

= .0165

= 0.12845

If r1,2 = -.60

σp = (.5) 2 (.10) 2 + (.5) 2 (.20) 2 + 2(.5)(.5)(.10)(.20)(−.60)

E x p e c te d
=R u r n 1+7 ..01
e .t0025 5 % + (−.006) X X
= .0065
0
= .08062 8 .0 6 % 1 2 .8 5 % R is k (S ta n d a rd d e v ia tio n )

The negative correlation coefficient reduces risk without sacrificing return.

5. For all values of r1,2:

7-5
 E(Rport) = (.6 x .10) + (.4 x .15) = .12

σ port = (.6) 2 (.03) 2 + (.4) 2 (.05) 2 + 2(.6)(.4)(.03)(.05)(r1,2 )

= .000324 + .0004 + .00072(r1,2 )

= .000724 + .00072(r1,2 )

5(a).
.000724 + .00072(1.0) = .001444 = .0380

5(b).

.000724 + .00072(.75) = .001264 = .0356

5(c).
.000724 + .00072(.25) = .000904 = .0301

5(d).

.000724 + .00072(.00) = .000724 = .0269

5(e).
.000724 + .00072( −.25) = .000544 = .0233

5(f).
.000724 + .00072( −.75) = .000184 = .0136

5(g).
.000724 + .00072( −1.0) = .000004 = .0020

6(a). E(Rp) = (1.00 x .12) + (.00 x .16) = .12

σ p = (1.00) 2 (.04) 2 + (.00) 2 (.06) 2 + 2(1.00)(.00)(.04)(.06)(.70)

= .0016 + 0 + 0 = .0016 = .04

6(b). E(Rp) = (.75 x .12) + (.25 x .16) = .13

σ p = (.75) 2 (.04) 2 + (.25) 2 (.06) 2 + 2(.75)(.25)(.04)(.06)(.70)

= .0009 + .000225 + .00063 = .001755 = .0419

7-6
6(c). E(Rp) = (.50 x .12) + (.50 x .16) = .14

σ p = (.50) 2 (.04) 2 + (.50) 2 (.06) 2 + 2(.50)(.50)(.04)(.06)(.70)

= .0004 + .0009 + .00084 = .00214 = .0463

6(d). E(Rp) = (.25 x .12) + (.75 x .16) = .15

σ p = (.25) 2 (.04) 2 + (.75) 2 (.06) 2 + 2(.25)(.75)(.04)(.06)(.70)

= .0001 + .002025 + .00063 = .002755 = .0525

6(e). E(Rp) = (.05 x .12) + (.95 x .16) = .158

σ p = (.50) 2 (.04) 2 + (.95) 2 (.06) 2 + 2(.05)(.95)(.04)(.06)(.70)

= .000004 + .003249 + .00015960 = .0034126 = .0584

7-7
7. DJIA S&P Russell Nikkei
Month (R1) (R2) (R3) (R4) R1-E(R1) R2-E(R2) R3-E(R3) R4-E(R4)
1 .03 .02 .04 .04 .01667 .00333 .01333 .00833
2 .07 .06 .10 -.02 .05667 .04333 .07333 -.05167
3 -.02 -.01 -.04 .07 -.03333 -.02667 -.06667 .03883
4 .01 .03 .03 .02 -.00333 .01333 .00333 -.01167
5 .05 .04 .11 .02 .03667 .02333 .08333 -.01167
6 -.06 -.04 -.08 .06 -.07333 -.05667 -.10667 .02833
Sum .08 .10 .16 .19

7(a).
.08 .10
E(R 1 ) = = .01333 E(R 2 ) = = .01667
6 6
.16 .19
E(R 3 ) = = .02667 E(R 4 ) = = .03167
6 6

7(b). Σ1 = (.01667)2+ (.05667)2+ (-.03333)2+ (-.00333)2+ (.03667)2 + (-.07333)2

= .00028 + .00321 + .00111 + .00001 + .00134 + .00538 = .01133

σ 12 = .01133/5 = .00226

σ1 = (.00226)1/2 = .0476

Σ2 = (-.00333)2 + (.04333)2 + (-.02667)2 + (.01333)2 + (.02333)2 + (-.05667)2

= .00001 + .00188 + .00071 + .00018 + .00054 + .00321 = .00653

σ 22 = .00653/5 = .01306

σ2 = (.01306)1/2 = .0361

Σ3 = (.01333)2 + (.07333)2 + (-.06667)2 + (.00333)2 + (.08333)2 + (-.106672)2

= .00018 + .00538 + .00444 + .00001 + .00694 + .01138 = .02833

σ 32 = .02833/5 = .00567

σ3 = (.00567) 1/2 = .0753

Σ4 = (.00833)2+(-.05167)2+ (.03833)2+ (-.01167)2+(-.01167)2 + (.02833)2

= .00007 + .00267 + .00147 + .00014 + .00014 .00080 = .00529

σ =.00529/5 =.001058
2
4

7-8
 σ4 = (.001058)1/2 = .0325
7(c).
.00006 + .00246 + .00089 - .00004 + .00086 + .00416
COV1,2 =
5
= .00839/5 = .001678
.00004 + .00318 + .00178 + .00004 + .00194 + .00604
COV2,3 =
5
= .01302/5 = .002604
.00003 - .00224 - .00102 - .00016 - .00027 - .00161
COV2,4 =
5
= - .00527/5 = - .001054
.00011 - .00379 - .00256 - .00004 - .00097 - .00302
COV3,4 =
5
= - .01027/5 = - .002054

7(d). Correlation equals the covariance divided by each standard deviation.

Correlation (DJIA, S&P) = 0.001678/ [(0.0476)(0.0361)] = .9765
Correlation (S&P, R2000) = 0.002604/ [(0.0361)(0.0753)] = .9579
Correlation (S&P, Nikkei) = -0.001054/ [(0.0361)(0.0325)] = -0.8984
Correlation (R2000, Nikkei) = -0.002054/ [(0.0753)(0.0325)] =-0.8393

7(e).

σ 2,3 = (.5) 2 (.0361) 2 + (.5) 2 (.0753) 2 + 2(.5)(.5)(.002604)

= .05518
E(R) 2,3 = (.5)(.01667) + (.5)(.02667) = .02167
σ 2,4 = (.5) 2 (.0361) 2 + (.5) 2 (.0325) 2 + 2(.5)(.5)(−.001054)
= .009875
E(R) 2,4 = (.5)(.01667) + (.5)(.03167) = .02417

The resulting correlation coefficients suggest a strong positive correlation in returns for
the S&P 500 and the Russell 2000 combinations (.96), preventing any meaningful
reduction in risk (.05518) when they are combined. Since the S&P 500 and Nikkei have a
negative correlation (-.90), their combination results in a lower standard deviation
(.009875).

8.
Cov i, j 100 100
ri, j = = = = 0.3759
σ iσ j 19 x 14 266

7-9
 APPENDIX 7

Answers to Problems

Appendix A

1(a). When E(σ1) = E(σ2), the problem can be solved by substitution,

E(σ 1 ) 2 - r1,2 E(σ 1 ) E(σ 1 )

W1 =
E(σ 1 ) 2 + E(σ 1 ) 2 - 2 r1,2 E(σ 1 ) E(σ 1 )

so that

E(σ 1 ) 2 [1 - r1,2 ]
W1 =
2 E(σ 1 ) 2 - 2 r1,2 E(σ 1 ) 2

E(σ 1 ) 2 [1 - r1,2 ]
=
2 E(σ 1 ) 2 [1 - r1,2 ]

= 1/2 = .5

1(b).
(.06) 2 − (.5)(.04)(.06)
W1 =
(.04) 2 + (.06) 2 − 2(.5)(.04)(.06)
.0036 − .0012
=
.0016 + .0036 − .0024
.0024
= = .8571 (or 6/7)
.0028

Appendix B

Variance of the portfolio is zero when:

E(σ 2 ) .06
w1 = = = .6 w 2 = 1 - w 1 = 1 - .6 = .4
E(σ 1 ) + E(σ 2 ) .04 + .06

7 - 10