Tests of equal forecast accuracy and encompassing for nested models

We examine the asymptotic and finite-sample properties of tests for equal f orecast accuracy and encompassing applied to 1-step ahead forecasts from ne sted linear models. We first derive the asymptotic distributions of two sta ndard tests and one new test of encompassing and provide tables of asymptot ically valid critical values. Monte Carlo methods are then used to evaluate the size and power of tests of equal forecast accuracy and encompassing. T he simulations indicate that post-sample tests cats be reasonably well size d. Of the post-sample tests considered, the encompassing test proposed in t his paper is the most powerful. We conclude with an empirical application r egarding the predictive content of unemployment for inflation. (C) 2001 Els evier Science S.A. All rights reserved.