Week 6 Tutorial Solutions

Topic: The Risk and Return from Investing

Section: PROBLEMS (p.167)

6.1. Calculating return (R) for these assets:

(a) Rps = (Dt + (PE - PB)) / PB

where Dt = the preferred dividend

PE = ending price or sale price
PB = beginning price or purchase price

R = (5 + -7) / 70
= -2.86%

(b) Rw = (Ct + PC) / PB

where Ct is any cash payments paid (there are none for a warrant)

PC = price change during the period

R = (0 + 2)/11

= 18.18% for the three-month period

(c) Rb = (It + PC) / PB

= (240* + 60) / 870

= 34.5% for the two-year period.

*two-years of interest is assumed to be received (12% x 2 x $1000) at the end of two

years, or $240.

Calculating Return Relatives (RRs) for these examples:

(a) a R of -2.86% is equal to a RR of 0.9714 or (1.0+ [-0.0286])

(b) a R of 18.18% is equal to a RR of 1.1818

(c) a R of 34.5% is equal to a RR of 1.345

6.2. Returns for the S&P 500 for 2000-2002 are -9.05%, -11.85%, and -22.10%. For 2003
the return was 28.37%.

The arithmetic mean for 2000-2002 was -14.33%; the geometric mean, -14.52%.

The arithmetic mean for 2000-2003 was -3.6575%; the geometric mean, -5.38%.

1
6.3. $100(1.3148)(0.95153)(1.20367)(1.22312)(1.05966)(1.31057)(1.18539)(1.05665)
(1.16339) (1.31229) = 4.89140 = the cumulative wealth index for this period.

(4.89140)1/10 = 1.17204

1.17204 - 1.0 = 0.17204 or 17.204%

6.4. The geometric mean for the S&P 500 was 10% for 1926-2018, a period of 93 years.
Raise 1.10 to the 93rd power to obtain 7,701.63
The value of the trust would be $100,000 × 7,701.63 = $707,163,310.

6.5. The geometric mean for small stocks was 11.8%.

Raise 1.118 to the 93rd power to obtain 31,995.414
The value of the trust would be $100,000 × 31,995.414 = $3,199,541,372.

6.6. The geometric mean for long-term Treasury bonds was 5.5%.
Raise 1.055 to the 93rd power to obtain 145.3713
The value of the trust would be $100,000 × 145.3713 = $14,537,132.

6.7. The geometric mean for Treasury bills was 3.3%

Raise 1.033 to the 93rd power to obtain 20.48.
The value of the trust would be $100,000 × 20.48 = $2,048,000.

6.8. The geometric mean for corporates is 5.9%. 1.059 raised to the 93rd power =
206.6923. Therefore, cumulative wealth per dollar invested over this period was
$206.6923.

6.9. Given a cumulative wealth of $20.48 for Treasury bills for 1926-2018, the geometric
mean would be 3.3%. This is determined by taking the 93rd root of $20.48 and
subtracting 1.0.

6.10. Raise the inflation rate (as a decimal) + 1.0, which is 1.03, to the 93rd power to obtain
15.6265. Do the same for corporate bonds, raising 5.9% to the 93rd power to obtain
206.6923. Divide 206.6923 by 15.6265 to obtain 13.227.

6.11. Divide 72 by 3 to obtain 24 years.

6.12. Divide ending value by beginning value to obtain 15.63, and take the 93rd root to
obtain
3%.

6.13. Raise 1.055 to the 100th power to obtain $211.47. This would be the cumulative
wealth per dollar invested.

6.14. The two ways to calculate inflation-adjusted cumulative wealth:

1. 1.055 / 1.03 = 1.02427; (1.02427)93 = 9.3029

2. (1.055)93 = 145.3713; (1.03)93 = 15.6265; 145.3713 / 15.6265 = 9.3029

2
6.15. The 6 Rs for 1926-1931 are, in order starting with 1926: 8.2%, 32.76%, 38.14%,
-10.18%, -26.48%, and -43.49%. Converting to RRs, we have: 1.082, 1.3276, 1.3814,
0.8982, 0.7352, and 0.5651. Multiply these 6 RRs together to obtain 0.7405. Take
the 6th root and subtract the 1. 0 to obtain 0.9512. Subtract 0.9512 from 1.0 to convert
to a decimal geometric mean of -0.0488, or -4.88%.

Of course, if we already know that the ending wealth index (cumulative wealth index)
for the six years 1926-1931 is 0.7405, we can calculate the geometric mean by taking the
sixth root of this wealth index and subtracting out the 1.0 to obtain the same answer
as above, -0.0488, or -4.88%.

6.16. Any set of Rs that are identical will produce a geometric mean equal to the arithmetic
mean; for example, 10%, 10% and 10%, or any other set of three identical numbers.

6.17. The calculated results for 1981-1991 are:

Arithmetic Mean 15.77%

Standard Deviation 13.15%
Geometric Mean 15.07%

As we can see, the standard deviation for the shorter period was less than that of the
entire period. This is because of the good years in the 1980s and 1990s that were
more similar than in a typical 20-year period. Also, there were only two negative
years during this period, whereas the historical norm for many years was 3 negative
years out of 10.

6.18. A compound rate of return of 10.4% for 10 years has a cumulative effect of 1.104
raised to the 10th power, or 2.6896. Therefore, $20,000 would grow to $20,000 ×
2.6896 =
$53,792. Thus, the $20,000 portfolio would be the better alternative.

sSection: COMPUTATIONAL PROBLEMS (p. 168)

6.1. First, convert the Rs to Return Relatives: 1.1506, 1.099, 1.161, 1.197, and 1.107.
Multiply these RRs together to obtain 1.94534, the cumulative wealth for the first 5
years.

A geometric mean of 10% for the decade results in a cumulative wealth of (1.10)10 =
2.594. Divide 2.594 by 1.94534 to obtain 1.3334. The fifth root of 1.3334 is 5.92%.
Thus, the geometric mean for the last 5 years must be 5.92% if the entire decade is to
have a geometric mean of 10%.

6.2. Cumulative wealth for the first 5 years is 1.94534 (from Computational Problem 6.1).
Cumulative wealth for 10 years, given a geometric mean of 10%, = (1.10)10 = 2.594.
If one of the next 5 years has a loss of 10%, the cumulative wealth for 6 years would
be 1.94534 (0.9) = 1.7508.

Therefore, divide 2.594 by 1.7508 to obtain 1.4816.

Take the 4th root of 1.4816 to obtain 1.1033; subtract 1.0 to obtain 10.33%.

3
 Therefore, the geometric mean of the remaining 4 years must be 10.33% for the
decade to show a geometric mean of 10%.

6.3. Raise 3.3% to the 93rd power to obtain $20.48. Divide $20.48 by 15.63 to obtain 1.31
take the 93rd root and subtract 1 to obtain 0.00291 or 0.291%.

6-4. Convert Returns to Return Relatives to obtain 1.093, 0.938, 1.121, and 1.074.
Multiply these Return Relatives together to obtain $1.234. Because this is equal to
the cumulative wealth for all five years, the Return for 2010 must be zero (a Return
Relative of 1.0).

4
 Topic: Portfolio Theory

Solutions to selected questions

Section: Questions (p. 193)
7.23. The expected return is 0.75 x E(return on stocks) + 0.25 x E(return on bonds)

=0.75(15) + 0.25(5)

=12.5 percent

The standard deviation is

σ = [w2stocks σ2stocks + w2bonds σ2bonds + 2wstockswbonds

Corr(Rstocks,Rbonds) σstocks σbonds]1/2

= [0.752 (225) + 0.252 (100) + 2(0.75)(0.25)(0.5)(15)(10)]1/2

= (126.5625 + 6.25 + 28.125)1/2
= (160.9375)1/2
= 12.69%

7.24. Define

Rp = return on the portfolio

R1 = return on the risk-free asset
R2 = return on the risky asset
w1 = fraction of the portfolio invested in the risk-free asset
w2 = fraction of the portfolio invested in the risky asset

Then the expected return on the portfolio is

E(Rp) = w1E(R1) + w2E(R2)

= 0.10(5%) + 0.9 (13%) = 0.5+11.7 +12.2%

To calculate standard deviation of return, we calculate variance of return and take the
square root of variance:

σ 2 (Rp) = w21 σ2 (R1) + w22 σ2 (R2) + 2w1w2Cov(R1,R2)

= 0.12(02) + 0.92 (232) + 2(0.1)(0.9)(0)
= 0.92 (232)
= 428.49

Thus, the portfolio standard deviation of return is σ (Rp) = (428.49)1/2 = 20.7 percent.

7.25. No—their systematic risk differs, and they should be priced in relation to their
systematic risk. This will be discussed in Chapter 9.

7.26. c (portfolio expected return depends only on the weights and security expected
returns)

5
7.27. d (note: for answer b, expected return is always a weighted average)

7.28. c (30 securities would have 30 x 30 = 900 terms)

7.29. a, b, d (c is incorrect; the portfolio variance equation does not include expected return)

Section PROBLEMS (p. 193-194)

7.1. (.15)(.20) = .030
(.20)(.16) = .032
(.40)(.12) = .048
(.10)(.05) = .005
(.15)(-.05) = -.0075
.1075 or 10.75% = expected return

To calculate the standard deviation, use the formula

n
VARi = Σ [Ri-E(Ri)]2Pi
i=1

VARGF = [(.20-.1075)2.15] + [(.16-.1075)2.20] +

[(.12-.1075)2.40] + [(.05-.1075)2.10]
+ [(-.05-.1075)2.15]

= .00128 + .00055 + .00006 + .00033 + .00372

= 0.00594

Since σi = (VAR)1/2
the σ for GF = (0.00594)1/2 = 0.0771 = 7.71%

7.2. (a) (.25)(15) + (.25)(12) + (.25)(30) + (.25)(22) = 19.75%

(b) (.10)(15) + (.30)(12) + (.30)(30) + (.30)(22) = 20.70%

(c) (.10)(15) + (.10)(12) + (.40)(30) + (.40)(22) = 23.50%

7.3. (a) (1) (1/3)2(10)2 = 11.089

+ (1/3)2( 8)2 = 7.097
+ (1/3)2(20)2 = 44.360
+ (2)(1/3)(1/3)(.6)( 8)(10) = 10.645
+ (2)(1/3)(1/3)(.2)(20)(10) = 8.871
+ (2)(1/3)(1/3)(-1)(20)( 8) = -35.485
46.577
variance = 46.577; σ = 6.82%

(2) variance = (.5)2(8)2 + (.5)2(20)2 + 2(.5)(.5)

(-1)(20)(8)

6
 = 16 + 100 - 80
= 36
σ = 6%

(3) variance = (.5)2(8)2 + (.5)2(16)2 +

2(.5)(.5)(.3)(8)(16)
= 16 + 64 + 19.2
= 99.2
σ = 9.96%

(4) variance = (.5)2(20)2 + (.5)2(16)2 +

2(.5)(.5)(0.8)(20)(16)
= 100 + 64 + 128
= 292
σ = 17.09%

(b) (1) variance = (.4)2(8)2 + (.6)2(20)2 + 2(.6)(.4)

(-1)(8)(20)
= 10.24 + 144 - 76.8
= 77.44
σ = 8.8%

(2) variance = (.6)2(8)2 + (.4)2(20)2 + 2(.6)(.4)

(-1)(8)(20)
= 23.04 + 64 - 76.8
= 10.24
σ = 3.2%

(c) In part (a), the minimum risk portfolio is 50% of the portfolio in B and 50% in
C. But this may not be the highest return. For the combinations in (a) above, the
risk/return combinations are:

Portfolio E(R) σ___

(1) ABC 19% 6.82%
(2) BC 21% 6.00%
(3) BD 17% 9.96%
(4) CD 26% 17.09%

Combination BC is clearly preferable over ABC and BD because there is a

higher E(R) at lower risk. The choice between BC and CD would
depend on the investor's risk-return tradeoff preference.

Section: COMPUTATIONAL PROBLEMS (p. 194)

7.1. The expected return for the third case shown in the table-- 0.6 weight on EG&G and
0.4 weight on GF is shown below. Each of the other expected returns in column 1 is
calculated exactly the same way.

7
 E(Rp) = 0.6 (25) + 0.4 (23) = 24.2%

7.2. The portfolio variance for the third case, with 0.6 weight on EG&G and 0.4 weight on
GF is shown below. Each of the other portfolio variances in column 2 is calculated
exactly the same way.

variancep = (.6)2(30)2 + (.4)2(25)2 + 2(.6)(.4)(112.5)

= 324 + 100 + 54
= 478

7.3. Knowing the variance for any combination of portfolio weights, the standard
deviation is simply the square root. Thus, for the case of 0.6 and 0.4 weights using
the variance calculated in Problem 7.2, we confirm the standard deviation as

(478)1/2 = 21.86 or 21.9 as per column 3.

7.4. The lowest risk portfolio would consist of 20% in EG&G and 80% in GF.

7.5. (a) E(R) = (.6)(7) + (.4)(12) = 9%; σ = .4(21) = 8.4%

(b) E(R) = (-.5)(7) + (1.5)(12) = 14.5%; σ = 1.5(21) = 31.5%

(c) E(R) = (0)(7) + (1.0)(12) = 12%; σ = 21%