Figure A.4:
A diagram of a line graph shows Price of Risk Lambda Estimation without Intercept, T equals 200 and N equals 31, comparing probability to reject lambda equals 0 versus population correlation with true factor.
Description: This figure complements Figure 5 and increases T to 200 years.
Interpretation: The Fama-MacBeth/Shanken approach slightly over-rejects “useless” factors in large samples. The bootstrap confidence intervals do not over-reject “useless” factors in large samples.

FMB-Shanken and GRS-FAR Power Curves With Varying the Risk Factor (Betas): Misspecified Model, T = 200.

This Feature Is Available To Subscribers Only

or Create an Account

Close subscription notice
Close access options