Finite sample efficiency of OLS in linear regression models with long-memory disturbances

Abstract

OLS is as efficient as GLS in the linear regression model with long-memory errors as the long-memory parameter approaches the boundary of the stationarity region, provided the model contains a constant term. This generalizes previous results of Samarov & Taqqu (Journal of Time Series Analysis 9, 1988, pp. 191-200) to the regression case and gives a further example of the `high-correlation asymptotics' of Kr?amer & Baltagi (Economics Letters 50, 1996, pp. 13-17).