Monte Carlo comparison of model and moment selection and classical inference approaches to break detection in panel data models

Abstract

Abstract This note shows how to use the GMM model and moment selection procedures of [Andrews, D. W. K., and B. Lu (2001): “Consistent model and moment selection procedures for GMM estimation with application to dynamic panel data models,” Journal of Econometrics, 101, 123–164.] for the purpose of detection of a general structural break in a dynamic panel data model. It compares the resulting method with the classical hypothesis testing approach of [De Wachter, S., and E. Tzavalis (2004): “Detection of structural breaks in linear dynamic panel data models,” QM University of London working paper nr 505.]. Out of 3 model selection criteria studied, the GMM-HQIC criterion is found to perform most similarly to the classical hypothesis test.