Finite sample efficiency of OLS in linear regression models with long-memory disturbances (Q5941374)

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 \textit{A. Samarov} and \textit{M. S. Taqqu} [J. Time Ser. Anal. 9, 191-200 (1988; Zbl 0637.62085)] and gives a further example of the `high-correlation asymptotics' of \textit{W. Krämer} and \textit{B. Baltagi} [Econ. Lett. 50, 13-17 (1996; Zbl 0900.90161)].