Model Misspecification - Schweser question Reading #10

N

NOSCIENCE

New member
Joined
Jan 2, 2015
Messages
0
Reaction score
0
This question is from reading #10, question 12 of Schweser study notes.
Which of the following situation is least likely to result in the misspecification of a regression model with model with monthly returns as the dependent variable?
A. Failing to include an idependient variable athat is related to monthy returns
B. Using leading P/E from the previous period as an idependent variable.
C. Using actual inflation as an diependent variable to proxy for expected inflation
Answer is B. My question is how can option B NOT result in model misspecification? Misspecification #4 says Using a Lagged Dependent Variable as an Idependent Variable (page 193 Schweser). Wouldn’t using P/E from the previous period as an independent variable be an example of this misspecification? Can anyone clarify?
 
NOSCIENCE wrote:
This question is from reading #10, question 12 of Schweser study notes.
Which of the following situation is least likely to result in the misspecification of a regression model with model with monthly returns as the dependent variable?
A. Failing to include an idependient variable athat is related to monthy returns
B. Using leading P/E from the previous period as an idependent variable.
C. Using actual inflation as an diependent variable to proxy for expected inflation
Answer is B. My question is how can option B NOT result in model misspecification? Misspecification #4 says Using a Lagged Dependent Variable as an Idependent Variable (page 193 Schweser). Wouldn’t using P/E from the previous period as an independent variable be an example of this misspecification? Can anyone clarify?
Click to expand...
If you omit a variable that is in the population (true) regression, you’ve misspecified the model. If you use a proxy variable, it is possible that the model is misspecified, since the actual (intended, and necessary) variable isn’t being used. As for choice “B”, they didn’t say that the regression uses leading P/E as the DV, so they’re not using a lagged DV as a regressor (monthly returns are the DV). Choices “A” and “C” are most likely to cause a misspecification issue; choice “B” is least likely.
 
You must log in or register to reply here.
Back
Top