On the Estimation of a Linear Time Trend Regression with a One-Way Error Component Model in the Presence of Serially Correlated Errors

In this paper we study the limiting distributions for ordinary least squares (OLS), fixed effects (FE), first difference (FD), and generalized least squares (GLS) estimators in a linear time trend regression with a one-way error component model in the presence of serially correlated errors. We show that when the error term is I(0), the FE is asymptotically equivalent to the GLS. However, when the error term is I(1) the GLS could be less efficient than the FD or FE estimators, and the FD is the most efficient estimator. However, when the intercept is included in the model and the error term is I(0), the OLS, FE, and GLS are asymptotically equivalent. Monte Carlo experiments are employed to compare the performance of these estimators in finite samples. The main findings are (1) the two-step GLS estimators perform well if the variance component, delta, is small and close to zero when rho < 1; (2) the FD estimator dominates the other estimators when rho = 1 for all values of delta; and (3) the FE estimator is recommended in practice since it performs well for all values of rho and delta.