A power comparison among tests for time reversibility

Since time reversibility (TR) is a necessary condition for an independent and identically distributed (iid) sequence, several tests for TR have been suggested to be applied as tests for model misspecification. In this paper, we compare the power of two well known TR tests against two situations: 1) the fitted model is a linear ARMA when the true data generating process is a nonlinear−in−mean model (either threshold autorregresive or bilinear), and 2) the fitted model is a symmetric GARCH model but the true process belongs to the asymmetric GARCH family (either EGARCH or GJR). The authors acknowledge financial support from the Ministerio de Ciencia y Tecnología, project SEC2003−09205. The authors also acknowledge the many useful comments of the Associate Editor Timothy Conley