Etf2480
Practice Questions and Past Exam Questions Question 1: Short answer questions. (a). It is widely accepted that empirical distributions of returns of various financial assets exhibit leptokurtosis. Explain what leptokurtosis is and how one can verify whether a certain empirical distribution is leptokurtic or not. (b). Fama-French three factor model attempts to explain the excess return on a certain asset in terms of three explanatory variables. This model includes two additional explanatory variables to the standard CAPM. Specify the Fama-French Model and explain what these two additional explanatory variables are and what empirical evidence that Fama and French used to justify these two additional explanatory variables in the model. (c). Explain the problem of a dummy variable trap.
(d). Discuss the consequences of including an irrelevant variable.
Question 2: Consider the following CAPM model.
where Rgt, Rft, and Rmt denote the returns on Gold, the risk free asset, and the market portfolio respectively. We use the Treasury bill as the risk free rate. It is converted to the monthly rate from the annual rate by dividing the raw data with 12. The returns are NOT expressed in percentages. We collect monthly data of Gold prices, the annual rate of the Treasury bill, and S&P 500 index from Jan. 1991 to Sep. 1999. The OLS estimates of the above regression are reported in the following Table 2.1.
Dependent Variable: DLOG(GOLD)-TBILL/12 Sample (adjusted): 1990M05 1999M09 Included observations: 113 after adjustments Variable C DLOG(SP500)-TBILL/12 R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient -0.006314 -0.087523 0.018504 0.009662 0.024662 0.067509 259.0527 2.092671 0.150825 Std. Error 0.002370 0.060502 t-Statistic -2.663494 -1.446607 Prob. 0.0089 0.1508 -0.007017 0.024782 -4.549605 -4.501333 -4.530017 1.728580
Mean dependent var S.D. dependent var Akaike info criterion Schwarz
(d). Discuss the consequences of including an irrelevant variable.
Question 2: Consider the following CAPM model.
where Rgt, Rft, and Rmt denote the returns on Gold, the risk free asset, and the market portfolio respectively. We use the Treasury bill as the risk free rate. It is converted to the monthly rate from the annual rate by dividing the raw data with 12. The returns are NOT expressed in percentages. We collect monthly data of Gold prices, the annual rate of the Treasury bill, and S&P 500 index from Jan. 1991 to Sep. 1999. The OLS estimates of the above regression are reported in the following Table 2.1.
Dependent Variable: DLOG(GOLD)-TBILL/12 Sample (adjusted): 1990M05 1999M09 Included observations: 113 after adjustments Variable C DLOG(SP500)-TBILL/12 R-squared Adjusted R-squared S.E. of regression Sum squared resid Log likelihood F-statistic Prob(F-statistic) Coefficient -0.006314 -0.087523 0.018504 0.009662 0.024662 0.067509 259.0527 2.092671 0.150825 Std. Error 0.002370 0.060502 t-Statistic -2.663494 -1.446607 Prob. 0.0089 0.1508 -0.007017 0.024782 -4.549605 -4.501333 -4.530017 1.728580
Mean dependent var S.D. dependent var Akaike info criterion Schwarz