A0500
Title: Practical aspects of using quadratic moment conditions in linear dynamic panel data models
Authors: Markus Fritsch - University of Passau (Germany)
Andrew Adrian Yu Pua - Xiamen University (China) [presenting]
Joachim Schnurbus - University of Passau (Germany)
Abstract: The focus is on the estimation of the lag parameter of linear dynamic panel data models with first-order dynamics based on the quadratic moment conditions. We first show that extending the standard assumptions to allow for mean stationarity and time series homoscedasticity and employing these assumptions in estimation restores standard asymptotics and mitigates the non-standard asymptotic distributions found in the literature. Because using these additional assumptions for estimation purposes may be too restrictive for practical usage, we analyze theoretically an IV estimator based on the quadratic moment conditions and provide a practical data-based approach to detect whether one would face a data generating process that does not suffer from non-standard behavior, while maintaining the default no serial correlation assumption.
