CHAPTER 10: SINGLE-INDEX AND MULTIFACTOR MODELS

l. a. To optimize this portfolio one would need: n n = ________ = b. = = 1770 1890 60 estimates of means 60 estimates of variances estimates of covariances _______________________ estimates ri rf = i + i(rM rf) + ei Ri = i + i RM + ei

In a single index model: or equivalently, using exess returns

the variance of the rate of return on each stock can be decomposed into the components: (l) The variance due to the common market factor (2) 2(ei) The variance due to firm specific unanticipated events In this model Cov(ri,rj) = ij. The number of parameter estimates would be: n = 60 estimates of the mean E(ri), n = 60 estimates of the sensitivity coefficient i, n = 60 estimates of the firm-specific variance 2(ei), and 1 estimate of the market mean E(rM) 1 estimate for the market variance 182 estimates Thus, the single index model reduces the total number of required parameter estimates from 1,890 to 182, and in general from (n2 + 3n)/2 to 3n + 2. 2. a. The standard deviation of each individual stock is given by: i = [+ 2(ei) ]1/2 Since A = .8, B = 1.2, (eA) = 30%, (eB) = 40%, and M = 22% we get: A = (.82 222 + 302)1/2 = 34.78% 10-1

B = (1.22 222 + 402)1/2 = 47.93% b. The expected rate of return on a portfolio is the weighted average of the expected returns of the individual securities: E(rp) = wAE(rA) + wBE(rB) + wfrf where wA, wB, and wf are the portfolio weights of stock A, stock B, and T-bills, respectively. Substituting in the formula we get: E(rp) = .30 13 + .45 18 + .25 8 = 14% The beta of a portfolio is similarly a weighted average of the betas of the individual securities: P = wAA + wBB + wff The beta of T-bills (f ) is zero. The beta of the portfolio is therefore: P = .30 .8 + .45 1.2 + 0 = .78 The variance of this portfolio is : = + 2(eP) where is the systematic component and 2(eP) is the nonsystematic component. Since the residuals, ei are uncorrelated, the non-systematic variance is: 2(eP) = w2 (eA) +w2(eB) + w2(ef) = .302 302 + .452 402 + .252 0 = 405 where 2(eA) and 2(eB) are the firm-specific (nonsystematic) variances of stocks A and B, and 2(ef), the nonsystematic variance of T-bills, is zero. The residual standard deviation of the portfolio is thus: (eP) = (405)1/2 = 20.12% The total variance of the portfolio is then:

10-2

= .782 222 + 405 = 699.47 and the standard deviation is 26.45%. 3. a. The two figures depict the stocks' security characteristic lines (SCL). Stock A has a higher firm-specific risk because the deviations of the observations from the SCL are larger for A than for B. Deviations are measured by the vertical distance of each observation from the SCL. Beta is the slope of the SCL, which is the measure of systematic risk . Stock B's SCL is steeper, hence stock B's systematic risk is greater. The R2 (or squared correlation coefficient) of the SCL is the ratio of the explained variance of the stock's return to total variance, and the total variance is the sum of the explained variance plus the unexplained variance (the stock's residual variance). R2 = Since stock B's explained variance is higher (its explained variance is , which is greater since its beta is higher), and its residual variance 2(eB) is smaller, its R2 is higher than stock A's. d. Alpha is the intercept of the SCL with the expected return axis. Stock A has a small positive alpha whereas stock B has a negative alpha; hence stock A's alpha is larger. The correlation coefficient is simply the square root of R2, so stock Bs correlation with the market is higher. Firm-specific risk is measured by the residual standard deviation. Thus, stock A has more firm-specific risk: 10.3% > 9.1%. Market risk is measured by beta, the slope coefficient of the regression. A has a larger beta coefficient: 1.2 > .8. R2 measures the fraction of total variance of return explained by the market return. A's R2 is larger than B's: .576 > .436. The average rate of return in excess of that predicted by the CAPM is measured by alpha, the intercept of the SCL. A = 1% is larger than B = 2%. Rewriting the SCL equation in terms of total return (r) rather than excess return (R): rA rf = + (rM rf)

b.

c.

e.

4. a. b.

c.

d.

e.

10-3

rA = + rf(1 ) + rM The intercept is now equal to: + rf(1 ) = 1 + rf (l 1.2) Since rf = 6%, the intercept would be: 1 1.2 = .2%.

5.

The standard deviation of each stock can be derived from the following equation for R2: R= = Therefore, = = = 980 A = 31.30% For stock B = = 4800 B = 69.28%

6.

The systematic risk for A is = .702 202 = 196 and the firm-specific risk of A (the residual variance) is the difference between A's total risk and its systematic risk, 980 196 = 784 B's systematic risk is: = 1.22 202 = 576 and B's firm-specific risk (residual variance) is: 4800 576 = 4224

7.

The covariance between the returns of A and B is (since the residuals are assumed to be uncorrelated): 10-4

Cov(rA,rB) = A B = .70 1.2 400 = 336 The correlation coefficient between the returns of A and B is: AB = = = .155 Note that the correlation is the square root of R2: = Cov(rA,rM) = AM = .201/2 31.30 20 = 280 Cov(rB,rM) = BM = .121/2 69.28 20 = 480

8.

9.

The non-zero alphas from the regressions are inconsistent with the CAPM. The question is whether the alpha estimates reflect sampling errors or real mispricing. To test the hypothesis of whether the intercepts (3% for A, and 2% for B) are significantly different from zero, we would need to compute t-values for each intercept. For portfolio P we can compute: P = [.62 980 + .42 4800 + 2 .4 .6 336]1/2 = [1282.08]1/2 = 35.81% P = .6 .70 + .4 1.2 = .90 2(eP) = = 1282.08 .902 400 = 958.08 Cov(rP, rM) = P= .90 400 = 360 This same result can also be attained using the covariances of the individual stocks with the market: Cov(rP, rM) = Cov( .6rA+.4rB, rM) = .6 Cov(rA, rM) + .4 Cov(rB, rM) = .6 280 + .4 480 = 360

10.

11.

Note that the variance of T-bills and their covariance with any asset are zero. Therefore, for portfolio Q

10-5

Q = w+ w+ 2 wP wM Cov(rP, rM) Q = [.52 1282.08 + .32 400 + 2 .5 .3 360]1/2 = [464.52]1/2 = 21.55% Q = .5 .90 + .3 1 + 0 = .75 2(eQ) = = 464.52 .752 400 = 239.52 Cov(rQ,rM) = Q= .75 400 = 300

12.

In a two-stock capital market, the capitalization of A being twice that of B implies that wA = 2/3 and wB = 1/3. A = 30%, B = 50%, AB = .7 a. The variance of the market index portfolio is: = w+ w+ 2wAwB AB = (2/3)2302 + (1/3)2502 + 2(2/3)(1/3).7 30 50 = 1144.44 = 33.83% b. The beta of stock A is: A = Cov(rA,rM) / where Cov(rA,rM) = Cov[rA,(2/3 rA +1/3 rB)] = 2/3 + 1/3 Cov(rA, rB) = (2/3) 302 + (1/3) .7 30 50 = 950 so that A= = .83 For stock B, Cov(rB, rM) = Cov[ rB, (2/3 rA +1/3 rB) ] = 2/3 Cov(rA,rB) + 1/3 = 2/3 .7 30 50 + 1/3 502 = 1533.33 so that, B = = 1.34 c. The residual variance of each stock is: 10-6

2(eA) = = 302 (.832 1144.44) = 111.60 and 2(eB) = = 502 (1.342 1144.44) = 445.04 d. If the index model holds, then the following holds too: (rA rf ) = A(rM rf ) 11% = .83(rM rf ) Thus the market risk premium must be rM rf = 11%/.83 = 13.25% Since A's beta is smaller than 1.0, its risk premium is smaller than the market's risk premium.

10-7

13. a. Merrill Lynch adjusts beta by taking the sample estimate of beta and averaging it with 1.0, using the weights of 2/3 and 1/3, as follows: adjusted beta = (2/3) 1.24 + (1/3) 1 = 1.16 b. If you use your current estimate of beta to be t1 = 1.24, then t = .3 + .7 (1.24) = 1.168 which is the prediction of beta for next year.

14.

The regression results provide quantitative measures of return and risk based on monthly returns over the 1992-2001 period. ABC: for ABC was .60, considerably below the average stocks of 1.0, indicating that when the S&P 500 rose or fell by 1 percentage point, ABCs return on average rose or fell only 0.60 percentage point. As such it indicates that ABCs systematic risk or market risk was low relative to the typical value for stocks. ABC's alpha (the intercept of the regression) was 3.2%, indicating that when the market return was 0%, the average return on ABC was 3.2%. ABC's unsystematic or residual risk, as measured by (e), was 13.02%. Its R2 was .35, indicating closeness of fit to the linear regression above the value for a typical stock. XYZ: for XYZ was somewhat higher at .97, indicating XYZs return pattern was very similar to the market indexs of 1.0 and the stock therefore had average systematic risk over the period examined. Alpha for XYZ was positive and quite large, indicating an almost 7.3% return, on average, for XYZ independent of market return. Residual risk was 21.45%, half again as much as ABCs, indicating a wider scatter of observations around the regression line for XYZ. Correspondingly, the fit of the regression model was considerably less, consistent with an R2 of only .17. The effects of including one stock or the other in a diversified portfolio may be quite different, if it can be assumed that both stocks' betas will remain stable over time, since there is such a large difference in their systematic risk level. The betas obtained from the two brokerage houses may help the analyst draw inferences for the future. ABC's estimates are similar regardless of the sample period of the underlying data. They range from .60 to .71, all well below the market average of 1.0. XYZ's varies significantly among the three sources of calculations, ranging as high as 1.45 for the weekly price change observations over the most recent two years. One could infer that XYZ's beta for the future might be well above 1.0, meaning it may have somewhat more systematic risk than was implied by the monthly regression for the 1992 - 2001 period. 10-8

The upshot is that these stocks appear to have significantly different systematic risk characteristics. If these stocks are added to a diversified portfolio, XYZ will add more to total volatility.

15. a. The of stock A is: A = rA [rf + A(rM rf )] = 11 [6 + .8(12 6)] = .2% For stock B: B = 14 [6 + 1.5(12 6)] = 1% Stock A would be a good addition. A short position in B may be desirable. b. The reward to variability ratio of the stocks is : SA = = .5 SB = = .73 Stock B is superior when only one can be held. c. The R2 of the regression is .72 = .49, leaving 51% of total variance unexplained by the market, and therefore, interpreted as firm-specific. 9 = 3 + (11 3) implies that = .75.

16.

17. 18. 19.

b. d. b.

10-9