GENERALIZED LEAST SQUARES APPROACH FOR MOVING AVERAGE MODELS

摘要

A new Bayesian method for estimating moving average (MA) models is proposed. The proposed methodology is based on replacing lagged errors of the original MA model with appropriately lagged residuals from a long autoregression. Unlike Broemeling and Shaarawy (1988), the exact structure of the approximation error when replacing true errors with corresponding residuals is derived and used in deriving the posterior distribution of the model parameters. In addition, a modified version of Broemeling and Shaarawy's (1988) method is suggested. The proposed method and original Broemeling and Shaarawy's method and its modified version are compared using several simulation studies and a real data.