Chapter 13 – Empirical Evidence on Security Returns
13–14
34. Fama and French, in their 1992 study, found that
D. macroeconomic factors had better explanatory power than beta in describing portfolio
returns.
E. none of the above is true.
Difficulty: Moderate
35. Which of the following statements is true about models that attempt to measure the
empirical performance of the CAPM?
A. The conventional CAPM works better than the conditional CAPM with human capital.
B. The conventional CAPM works about the same as the conditional CAPM with human
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–15
36. Which of the following statements is false about models that attempt to measure the
empirical performance of the CAPM?
A. The conventional CAPM works better than the conditional CAPM with human capital.
Difficulty: Moderate
37. A study by Mehra and Prescott (1985) covered a period of ______ years and found that
historical average excess returns _________.
A. 25, have been too small to be consistent with rational security pricing.
B. 25, have been too large to be consistent with rational security pricing.
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–16
38. Fama and French (2002) studied the equity premium puzzle by breaking their sample into
subperiods and found that
A. the equity premium was largest throughout the entire 1872-1999 period.
B. the equity premium was largest during the 1872-1949 subperiod.
Difficulty: Moderate
39. Which of the following is (are) a result(s) of the Fama and French (2002) study of the
equity premium puzzle?
I) Average realized returns during 1950-1999 exceeded the internal rate of return (IRR) for
corporate investments.
II) The statistical precision of average historical returns is far higher than the precision of
estimates from the dividend-discount model (DDM).
III) The reward-to-variability ratio (Sharpe) ratio derived from the DDM is far more stable
than that derived from realized returns.
IV) There is no difference between DDM estimates and actual returns with regard to IRR,
statistical precision, or the Sharpe measure.
A. I, II, and III
Difficulty: Difficult
Chapter 13 – Empirical Evidence on Security Returns
13–17
40. Equity premium puzzle studies may be subject to survivorship bias because
A. the time period covered was not long enough.
B. an inappropriate index was used.
Difficulty: Moderate
41. Tests of the CAPM that use regression techniques are subject to inaccuracies because
A. the statistical results used are almost always incorrect.
Difficulty: Difficult
Chapter 13 – Empirical Evidence on Security Returns
13–18
42. Which of the following must be done to test the multifactor CAPM or the APT?
I) specify the risk factors
II) identify portfolios that hedge the risk factors
III) test the explanatory power of hedge portfolios
IV) test the risk premiums of hedge portfolios
A. I and II
B. II, and IV
Difficulty: Moderate
43. The Fama and French three factor model uses ___, ___, and ___ as factors.
A. industrial production, term spread, default spread
B. industrial production, inflation, default spread
Difficulty: Easy
Chapter 13 – Empirical Evidence on Security Returns
13–19
44. The Fama and French three factor model does not use ___ as one of the explanatory
factors.
A. industrial production
B. inflation
Difficulty: Easy
45. Davis, Fama, and French (2000) created three B/M ratio groups. The firms with a high
B/M ratio are often called
D. blend firms.
E. none of the above.
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–20
46. Jagannathan and Wang (2006) find that the CCAPM explains returns ______ the Fama-
French 3-factor model and that the the Fama-French 3-factor model explains returns ______
than the traditional CAPM.
A. worse than; worse than
B. worse than; better than
Difficulty: Moderate
47. A major finding by Heaton and Lucas (2000) is that:
A. the market rate of return does not help explain the rate of return of individual securities and
CAPM must be rejected.
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–21
48. Liew and Vassalou (2000) show that returns on style portfolios (SMB and HML)
A. seem like statistical flukes.
B. seem to predict GDP growth.
Difficulty: Moderate
49. Petkova and Zhang (2005) examine the relationship between beta and the market risk
premium and find
A. a countercyclical beta.
B. the beta of the HML portfolio is negative in good economies and positive in bad
economies.
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–22
50. Studies by Chan, Karceski, and Lakonishok (2003) and La Porta, Lakonishok, Shleifer,
and Vishny (1997) report that
D. all of the above
E. none of the above
Difficulty: Moderate
51. The Fama-French model
I) is a useful tool for benchmarking performance against a well-defined set of factors.
II) premia are determined by market irrationality.
III) premia are determined by rational risk factors.
IV) the reason for the premia is unsettled.
V) is not a useful tool for benchmarking performance against a well-defined set of factors.
A. I only.
B. V only
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–23
52. Acharya and Pedersen (2005) examine the effects of liquidity and find that
D. liquidity risk does not help explain size or book to market effects in assets returns.
E. none of the above.
Difficulty: Moderate
53. GARCH models are used to estimate
A. conditional average returns of stocks, indices, or portfolios.
B. unconditional average returns of stocks, indices, or portfolios.
Difficulty: Moderate
54. GARCH models were pioneered by
D. Michael Jensen
E. Fama and French
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–24
Short Answer Questions
55. Discuss the results of the studies of John Lintner (1965) and Merton Miller and Myron
Scholes (1972) in terms of the validity of the capital asset pricing model (CAPM).
These studies examined the returns of 631 NYSE listed stocks over the period, 1954 – 1963.
These results indicated that the SML was “too flat”, i.e., the g1 coefficient was too small and
the slope was too small, as predicted by the CAPM. These results were statistically
significant. In addition, nonsystematic risk, not systematic risk, appeared to predict expected
excess returns. However, these studies contained two methodological problems. First, the
Difficulty: Moderate
56. Discuss the Black Jensen Scholes (BJS) study of the zero-beta version of the CAPM.
Black, Jensen, and Scholes studied all available NYSE stock returns over the period 1931 to
1965. Their results indicated that the zero-beta version of the CAPM might better explain
observed market relationships than the simple version of the CAPM. When borrowing is
restricted, the CAPM expected return-beta relationship must be amended; that is, the risk-free
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–25
57. Discuss Roll’s critique of the CAPM.
In 1977, Richard Roll published a critique of the capital asset pricing model, in which he
argued not only that the empirical tests of the expected return-beta relationship are invalid, but
also that it is doubtful that the CAPM can be tested. These essence of Roll’s critique is as
follows: There is a single testable hypothesis associated with the CAPM, which is that the
market portfolio is mean-variance efficient; all other relationships of the model, including the
linear risk-return trade-off follow from the mean-variance efficiency tenet and are not
independently testable; in any sample of observations of individual returns, there will an
Difficulty: Moderate
Chapter 13 – Empirical Evidence on Security Returns
13–26
58. Describe some of the ways the CAPM is applied in practice.
The four items mentioned at the beginning of the chapter are
• Professional portfolio managers use the CAPM to determine appropriate security returns.
These managers also have their performances evaluated based on the reward-to-variability
Difficulty: Moderate
59. When portfolio performance is measured, what type of benchmark may be used? Explain
what Roll meant by benchmark error.
The benchmark portfolio should be broadly based since it is a proxy for the unobservable
market portfolio. Examples include, but are not limited to, the S&P500 Index, the NYSE
Composite Index, and the Wilshire 5000 Index.
Difficulty: Moderate