Predictive density criterion for SETAR models

Abstract

This paper deals with the problem of joint determination of the delay parameter and autoregressive orders of the Self-Exciting Threshold Autoregressive (SETAR) models. More specifically, we propose a variant of the Predictive Density Criterion (PDC) for the purpose of SETAR model selection. The performance of this variant is evaluated by means of Monte Carlo experiments. Our results indicate that PDC serves as an effective tool for jointly selecting the correct order and delay.

KEYWORDS:

• Bayesian approach
• Model selection problem
• Nonlinear time series
• Predictive density criterion
• SETAR model

Acknowledgments

The authors are deeply grateful to the Editor in Chief and the referee for their useful remarks and constructive suggestions that substantially improved the quality of the paper.