Problem Set 3
Problem Set 3
Problem Set 3
US Europe
Expected Return (in EUR) 0.12 0.14
EUR-Return Standard Deviation 0.10 0.13
a) Assume that the fund manager wants to maximize the utility of risk-averse fund
investors, and knows that fund investors can combine their investment in the fund with
the risk-free asset to achieve their desired optimal portfolio. What weights would he
choose?
b) How should the fund manager invest, if his investors are not able to combine the fund
with the risk-free asset, assuming that they have mu-sigma preferences, and they
maximize the following utility function with a risk aversion coefficient of 3:
U ( P , s P ) = P - g2 s P2
c) Assume that Volker is replaced by Robin, who uses more data to estimate expected
returns, and thinks that the expected returns in the US are slightly lower at 11% p.a.
How would the optimal portfolio under a) and b) look like in this case?
a) How can you test if the CAPM holds? What do Fama and French imply about the
empirical performance of the CAPM?
b) How can you reject the CAPM? Also, think of the Roll critique!
c) According to the theory of factor models, which variables could be factors? What is the
intuition behind firm characteristics being priced?
d) Give examples of other multifactor models and the factors they contain. Can you
propose a model that works both in, and out of sample at all times?
Problem 3: Multifactor models macro variables as factors. Discussion of the paper of
Chen, Roll and Ross (1986):
http://rady.ucsd.edu/faculty/directory/valkanov/pub/classes/mfe/docs/ChenRollRoss_JB_1986
.pdf
a) Chen, Roll and Ross (1986) examine whether innovation in macroeconomic variables
are priced in the stock markets. Why would the macroeconomic environment matter for
the stock markets?
b) If stock prices can be considered as the expected discounted dividend flows (Equations
1 and 2), then which are the two channels that can be affected by systematic forces?
c) Which variables do they look at, and what are the channels through which these
variables matter for stock prices?
d) Why do they add the market factor of the CAPM? Explain the mismatch in nature of
macroeconomic time series and stock prices.
In the following, use the dividend-discount-model to determine expected stock returns for each
of the two cases:
b) The stock just paid a dividend of 8. You expect the dividend to grow by 10% per year
in all future years. Determine the expected return of this stock, if the current stock price
is 180.
c) Assume that we are immediately before another stocks dividend payment date. The
dividend is 5 and you expect the dividend to stay constant in the future. The dividend
is paid semiannually. Determine the expected return of this stock, if the current stock
price is 190.
Problem 5: Fama and French 3-factor model
A fund manager uses the 3-factor model of Fama/French (1993) to determine the expected
returns of stocks A and B. Based on the historical returns of the two stocks up to now (time t),
he determines the following estimates for the alphas and betas:
Alpha Market-Beta SMB-Beta HML-Beta
Firm A 0.24% 1.20 0.31 -0.39
Firm B -0.22% 0.93 0.34 0.07
The one-year risk-free interest rate and the estimates for the market risk premium, the SMB-
factor and the HML-factor for the next year are (in %):
a) What are SMB and HML factors? How are they constructed?
b) Estimate the expected return of stock A and of stock B for the next year (t+1).
b) If Sarahs prediction model was perfect, what return would she earn over the 2009-2013
period with the strategy determined in a)?
c) Assume that the estimated constant from above is wrong. The true value is exactly zero.
What is the return lost to Sarah due to using the wrong model?
Problem 7: Discuss the momentum factor and mutual fund performance based on
Carhart (1997)
https://faculty.chicagobooth.edu/john.cochrane/teaching/35150_advanced_investments/
Carhart_funds_jf.pdf
a) What is momentum? How can one construct the momentum factor?
b) What is survivorship bias? Why is it important that you take care of this in your
analysis?
c) How do you construct a factor-mimicking portfolio for:
a. Momentum?
b. Liquidity?
d) According to the paper, what causes persistence in mutual fund performance? How does
fee structure matter?
e) How can one proxy for managerial skills? And how can one provide evidence that skill
exists? What is the general problem with these tests?