Figure A.2:
A diagram of a line graph shows GMM Estimation of Gamma, T equals 200 and N equals 1, comparing probability of finding a significant gamma versus population factor correlation.
Description: This figure complements Figure 3. T is increased to 200 years in the simulation.
Interpretation: The GMM-AR test has no power to detect “useful” factors in large samples. GMM-based standard errors over-reject “useless” factors but are more powerful in detecting “useful” factors. Bootstrap confidence intervals do not over-reject “useless” factors and are at the same time powerful in detecting “useful” factors.

The Power of GMM-Based Inference on the Coefficient of Relative Risk Aversion: T = 200.

This Feature Is Available To Subscribers Only

or Create an Account

Close subscription notice
Close access options