On the estimation and inference of a cointegrated regression in panel data

In this chapter we study the asymptotic distributions for ordinary least sq uares (OLS), fully modified OLS (FMOLS), and dynamic OLS (DOLS) estimators in cointegrated regression models in panel data. We show that the OLS: FMOL S, and DOLS estimators are all asymptotically normally distributed. However ; the asymptotic distribution of the OLS estimator is shown to have a non-z ero mean. Monte Carlo results illustrate the sampling behavior of the propo sed estimators and show that (I) the OLS estimator has a non-negligible bia s in finite samples, (2) the FMOLS estimator does not improve over the OLS estimator in general, and (3) the DOLS outperforms both the OLS and FMOLS e stimators.