- Mason is part of a team that has been charged by upper management to create a marketing program to present to both current and potential clients of ABC. He needs to be able to demonstrate a history of strong performance for the ABC funds, and, while not promising any measure of future performance, project possible return scenarios. He decides to conduct a regression analysis on all of ABC’s in-house funds. He is going to use 12 independent economic variables in order to predict each particular fund’s return. Mason is very aware of the many factors that could minimize the effectiveness of his regression model, and if any are present, he knows he must determine if any corrective actions are necessary. Mason is using a sample size of 121 monthly returns. In order to conduct an F-test, what would be the degrees of freedom used (dfnumerator; dfdenominator)? A) 108; 12. B) 120; 11. C) 12; 108. D) 11; 120. 2. In regards to multiple regression analysis, which of the following statements is most accurate? A) R2 is less than adjusted R2. B) Adjusted R2 is less than R2. C) Adjusted R2 always decreases as independent variables increase. D) Adjusted R2 does not account for the number of observations. 3. One of the most popular ways to correct heteroskedasticity is to: A) adjust the standard errors. B) employ residual plots. C) improve the specification of the model. D) use robust standard errors. T/G

- C 2. B 3. D (but I forget this one)

C B D

C B D EDIT: Nice, wee are in agreement.

nice job guys, i’ll have to put in harder ones 1. Your answer: C was correct! Degrees of freedom for the F-statistic is k for the numerator and n − (k + 1) for the denominator. k = 12 n − (k + 1) = 121 − (12 + 1) = 108 2. Your answer: B was correct! Whenever there is more than one independent variable, adjusted R2 is less than R2. Adding a new independent variable will increase R2, but may either increase or decrease adjusted R2. R2 adjusted = 1 − [((n − 1) / (n − k − 1)) × (1 − R2)] Where: n = number of observations K = number of independent variables R2 = unadjusted R2 3. Your answer: D was correct! Using generalized least squares and calculating robust standard errors are possible remedies for heteroskedasticity. Employing residual plots is a method to detect, not correct, heteroskedasticity. Improving specifications remedies serial correlation. The standard error cannot be adjusted, only the coefficient of the standard errors. T/G

C) C) This one is phrased kinda funny; do they mean Adjusted R2 gets smaller as more independent variables are added? Or do they mean it gets smaller as x increases in value? D) Just because you guys say so; really need to review quant this week and actually do some questions

Ohhh, I got burned on #2… I must’ve been thinking of R2 definitely decreasing as you add more independent variables. F@ck!

The answer for #3 is definitely D) but what about C)? Seems like GARCH models, etc. are about “correcting for heteroscedasticity”.