Simple procedures for testing autoregressive versus moving average errors in regression models

This paper develops several simple separate (or non-nested) procedures for testing autoregressive versus moving average errors in regression models. T hese asymptotically valid tests are straightforward to calculate: after est imating both models by maximum likelihood methods, the procedure involves t esting the significance of variables added to a linearized version of the n ull model, the added variables being the predictions, or the residuals from the specified alternative model, or the difference of the predictions of t he two models. Some small sample evidence on the properties of the tests is presented, as is an empirical application on the Australian unexpected inf lation rate series. JEL Classification Numbers: C12, C22, C52, E31.