Financial Economics Problem Set 3 Portfolio Theory Fall 2013 Problem 1 Consider two stocks, ABC and XYZ.

Z. The rates of return of the two stocks next year, rABC and rXYZ, depend on economic conditions, which are equally likely to be good, OK, or bad. rABC and rXYZ are expected to be as follows: Economic conditions Good OK Bad rABC 30.86% 18.96% -4.83% rXYZ 75.42% 11.10% -41.53%

a) Compute the expected rate of return and variance of returns of the two stocks. Compute the correlation coefficient between the rates of return of the two stocks. b) Suppose you want to invest $10,000 in portfolio P, which contains 50% of stock ABC and 50% of stock XYZ. Compute the expected rate of return and the variance of return of this portfolio. c) Suppose you can also invest in the stocks of utility companies. Each utility firm has an expected rate of return of 10% and a standard deviation of return of 5%. There is no correlation between the rates of return of two such firms. Assume you invest an equal amount in 15 such utility companies. Calculate the expected return and variance of return of this portfolio. Would increasing the number of utility companies in your portfolio affect the expected return and variance of the portfolio? Interpret your result. d) Now assume you invest a fraction ! of your wealth in the portfolio of part (c), and the rest of your wealth in portfolio P of part (b). Assume that the correlation between the rate of return of any utility firm and the rate of return of ABC or XYZ is zero. How should you choose ! if you want to minimize the variance of your wealth? Problem 2 The probability distribution for the returns of security X is: Probability Rate of return 0.1 60% 0.2 40% 0.4 30% 0.2 20% 0.1 -100%

There exists another risky security, Y, whose rate of return is linked to that of X by the following relation: rY = 0.06 + 0.2 rX. a) What is the coefficient of correlation between the returns of X and Y? b) Assume there also exists a risk-free security with a rate of return equal to 6%. Which profitable operation could you realize and under what conditions? Be sure to describe in details the operation you have in mind.

Problem 3 Consider an economy with two risky assets s1 and s2. Suppose that the expected rate of return of asset 1 is 10% and that !1=10%. The expected rate of return of asset 2 is 5% and !2=20%. The correlation coefficient between s1 and s2 is "1,2 = 0.875. a) What is the composition, the expected return and the standard deviation of the minimum variance portfolio? b) Represent in a graph all the risk/return combinations that can be obtained with portfolios containing s1 and s2. Clearly identify s1, s2 and the minimum variance portfolio. c) Harry is a mean-variance investor with risk aversion A=3. Harry is not allowed to short-sell asset s2 while he can short-sell asset s1. c.1) In a new graph, represent the risk/return combinations that are available to Harry. c.2) What is the composition of Harrys optimal portfolio? d) Let us introduce a risk free asset sf with return rf =8%. Now all short-sales are allowed: d.1) In a third graph represent, s1, s2 and sf , the set of risk/return combination that can be obtained with portfolios composed only of s1 and s2, the tangency portfolio T, and the capital allocation line. d.2) Portfolio D has the composition XD ={x1=1/3, x2=1/3, xf=1/3,}. Is portfolio D efficient? Explain carefully but concisely why it is or why it is not. d.3) What is the composition of Harrys optimal portfolio in this case (for this question, assume again that Harry cannot short-sell asset s2)?

Financial Economics Problem Set 3 Portfolio Theory Fall 2013 Problem 1 a) E(rABC)=1/3x30.86%+1/3x18.96%+1/3x-4.83%=15%. E(rXYZ)=1/3x75.42%+1/3x11.10%+1/3x-41.53%=15% Var(rABC)= E[(E(rABC)-rABC)"] =1/3x(15%-30.86%)"+1/3x(15%-18.96%)"+1/3x(15%+4.83%)"=2.20%. So #(rABC)=14.83%. Similarly, we find Var(rXYZ)=22.87% and #(rXYZ)=47.82%. Cov(rABC,rXYZ)=E(rABCxrXYZ)-E(rABC)xE(rXYZ) =1/3x30.86%x75.42%+1/3x11.10%x18.96%+1/3x-41.53%x-4.83%-15%x15%=6.88%. So $(rABC,rXYZ)= Cov(rABC,rXYZ)/(#(rABC)x#(rXYZ))=6.88%/(14.83%x47.82%)=97.01%. (b) Let rp be the return of the portfolio. E(rp)=1/2xE(rABC)+1/2xE(rXYZ)=1/2x15%+1/2x15%=15%. Var(rp)=Var(1/2xrABC+1/2xrXYZ)=(1/2)"xVar(rABC)+(1/2)"xVar(rXYZ)+2x1/2x1/2xCov(rABC, rXYZ)=9.71%. c) Let r1, r2, ., r15 be the returns of each of the 15 utilities. r is the return of the portfolio containing the 15 utilities. Each utility has a weight of 1/15 in this portfolio. E(r) = E(1/15xr1+1/15xr2++1/15xr15) = E(1/15x(r1+r2++r15))=1/15x15x10%=10%. Var(r) = Var(1/15xr1+1/15xr2++1/15xr15) = Var(1/15x(r1+r2++r15))=1/15"xVar(r1+r2++r15). =1/15"x[Var(r1)+Var(r2)++Var(r15)] = 1/15"x(15x5%x5%)=0.25%/15 = 0.02%. When the number of utilities in the portfolio increases, the expected return of the portfolio does not change and its variance decreases. If there are n such firms in the portfolio, its variance is equal to 0.25%/n. This is the consequence of diversification. d) Let Q be the new portfolio containing a weight ! of the portfolio of part (c), and a weight 1-! of portfolio P. Let rq be the return of this portfolio. Var(rq)=Var((1-!)rp+!r)=(1-!)"Var(rp)+ !"Var(r) =(!"-2!+1)Var(rp)+ !"x0.25%/15 =(Var(rp)+0.25%/15) !"-2!Var(rp)+Var(rp) Which reaches a minimum for: ! = 2Var(rp)/[2(Var(rp)+0.25%/15)] = Var(rp)/(Var(rp)+0.25%/15). Thus, ! = 9.71%/(9.71%+0.25%/15) = 0.9983. wp=1-!, therefore wABC=wXYZ=(1-!)/2 = 0.086%. Problem 2 a) The covariance between rX and rY is equal to cov(rX,0.06+0.2rX) = 0.2.cov(rX,rX) = 0.2.Var(rX). Therefore, the correlation coefficient between rX and rY is 0.2.Var(rX)/#X.#Y = 0.2Var(rX)/(#X.0.2.#X) = 1. b) E(RX) = 0.1#0.6 + 0.2#0.4 + 0.4#0.3 +0.2#0.2 + 0.1#(-1) = 20% E(RY) = 0.06 + 0.2#0.2 = 10% Let us look up if we can find a riskless portfolio. The return rate of a portfolio P composed of x of asset X and (1-x) of asset Y is RP = x rX + (1-x) rY = rX (x +(1-x)0.2) +(1-x)0.06. P is a riskless portfolio if (x +(1-x)0.2)=0, i.e. x = -0.25. As an alternative observe that: Var(rX) = 17.2% !(rX) = 41.47% !(rY) = 0.2 # !(rX) = 8.29% 3

The minimum variance portfolio is for x =

and (1-x) = 1.25 we get a risk-free portfolio. Hence, if short-sales are allowed one can construct a risk-free portfolio.

" (RY ) 0.0829 = = #0.25 . " (RY ) # " (RX ) 0.0829 # 0.4147

! portfolio = -0.25#0.2 + 1.25#0.1 = 7.5% Expected return on this risk-free This is larger than the rate of return of the risk-free security which is 0.06 $ there exists an arbitrage opportunity: borrow at the risk-free rate and invest in the synthetic riskless portfolio (-0,25 ; 1,25).
Problem 3 x1,MVP = 0.2% (0.2 0.875 % 0.1)/(0.22+0.12 2%0.875%0.1%0.2) = 1.5 x2,MVP = 0.5 E[r MVP] = 1.5%10% - 0.5%5% = 12.5% ! MVP =((1.5%0.1)2+(-0.5%0.2)2-2%1.5%0.5%0.875%0.2%0.1)1/2 =7.9% b) E[r]

MVP 10% 5% s1 s2

10%

20%

c) c.1) E[r]

10% 5%

s1 s2

c.2) Harry will invest all his wealth in asset 1 as this is the only efficient portfolio given his shortselling constraint. d) d.1) E[r] CAL

T MVP 10% 8% 5% s1 D s2

d.2) The tangency portfolio is obtained by short selling 2 and buying 1. An efficient portfolio is a combination of the risk-free asset and the tangency portfolio. Hence it must have a negative weight of asset 2 if it has a positive weight of asset 1. Portfolio D has positive weight of asset 1 and 2 hence it is not a combination of T and sf, It is not an efficient portfolio. (See picture)

d.3) E[r]

CAL 10% 8% 5% s1 s2

Now Harry can reach all points on the capital allocation line obtained by combining s1 and sf. The weight of asset s1 in his optimal portfolio verifies w1 = (10%-8%)/(3.10%2) = 2/3.