Functional Coefficient Autoregressive Models for Nonlinear Time Series

• Activity Number: 410
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Alireza Tahai and Mehrzad Netadj*+
• Companies: Mississippi State University
• Keywords: Linearity test ; Local linear estimation

Abstract

Constructing models from time series with nontrivial dynamics involves the problem of how to choose the best model from within a class of models, or to choose between competing classes. The Functional Coefficient Autoregressive (FAR) model is a rich class of models that includes many successful parametric nonlinear time series models such as the threshold Autoregressive models of Tong (1983), exponential AR models of Haggan and Ozaki (1978) and many others. The Taylor expansion of the exponential AR model around a given point in the sample space is used to examine the dynamic relationship of the model. The built model is close to the simplest possible according to a description length criterion.The method will deliver a linear model if that has shorter description length than a nonlinear model. The local linear regression is applied to estimate FAR model for time series data.