A simplified GLS estimator for autoregressive moving-average models

Abstract

Koreisha and Pukkila (1990a) have recently proposed a fast and efficient GLS estimator for the univariate ARMA time series model which appears to be far more robust than maximum likelihood methods and of comparable accuracy. The one drawback to this new estimator is that it requires use of the Cholesky decomposition. The purpose of this paper is to suggest an alternative simplified GLS estimator, which can be implemented with just repeated applications of an OLS subroutine. A limited Monte Carlo study establishes that this new estimator is just as efficient as that of Koreisha and Pukkila.