Asymptotic Inference for Dynamic Panel Estimators of Innite Order Autoregressive Processes

In this paper we consider the estimation of a dynamic panel autoregressive (AR) process of possibly innite order in the presence of individual effects. We utilize the sieve AR approximation with its lag order increasing with the sample size. We establish the consistency and asymptotic normality of the standard dynamic panel data estimators, including the xed effects estimator, the gen- eralized methods of moments estimator and Hayakawa's instrumental variables estimator, using double asymptotics under which both the cross-sectional sam- ple size and the length of time series tend to innity. We also propose a bias- corrected xed effects estimator based on the asymptotic result. Monte Carlo simulations demonstrate that the estimators perform well and the asymptotic approximation is useful. As an illustration, proposed methods are applied to dynamic panel estimation of the law of one price deviations among US cities.