Non-and Semiparametric Identiication of Seasonal Nonlinear Autoregression Models

Non-or semiparametric estimation and lag selection methods are proposed for three seasonal nonlinear autoregressive models of varying seasonal exibility. All procedures are based on either local constant or local linear estimation. For the semipara-metric models, after preliminary estimation of the seasonal parameters, the function estimation and lag selection are the same as nonparametric estimation and lag selection for standard models. A Monte Carlo study demonstrates good performance of all three methods. The semiparametric methods are applied to German real GNP and UK public investment data. For these series our procedures provide evidence of nonlinear dynamics. The authors thank Richard Blundell and Herman van Dijk for suggesting to investigate the seasonal shift model and JJ org Breitung and Helmut L utkepohl for valuable comments on an earlier draft. The helpful comments of two anonymous referees, the Associate Editor and the Editor are also very much appreciated.