Simulation Based Inference In Moving Average Models

We examine several autoregressive-based estimators for the parameters of a moving average process, including the estimators initially proposed by Galbraith and Zinde-Walsh [1994] and Gouriéroux, Monfort and Renault [1993]. We also propose over-identified asymptotic-least-squares based variants of the former, and extensions of the latter based on Gallant and Tauchen's [1996] simulated method of moments. The relative performance of these estimators is assessed, with emphasis on the near-uninvertibility region. We find that, although no formal local-to-one arguments are taken into consideration, the Wald-type indirect inference method performs best at the boundary, with practically just one calibration.