A generalized least squares estimation method for VARMA models. (Revised edition).

In this paper a new generalized least squares procedure for estimating VARMA models is proposed. This method differs from existing ones in explicitly considering the stochastic structure of the approximation error that arises when lagged innovations are replaced with lagged residuals obtained from a long VAR. Simulation results indicate that this method improves the accuracy of estimates with small and moderate sample sizes, and increases the frequency of identifying small nonzero parameters, with respect to both Double Regression and exact maximum likelihood estimation procedures.