A diagnostic statistic for functional-coefficient autoregressive models

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)

Abstract

In this paper, functional coefficient autoregressive (FAR) models proposed by Chen and Tsay (1993) are considered. We propose a diagnostic statistic for FAR models constructed by comparing between parametric and nonparametric estimators of the functional form of the FAR models. We show asymptotic properties of our statistic mathematically and it can be applied to the estimation of the delay parameter and the specification of the functional form of FAR models.

Original languageEnglish
Pages (from-to)2257-2273
Number of pages17
JournalCommunications in Statistics - Theory and Methods
Volume27
Issue number9
DOIs
Publication statusPublished - 1998
Externally publishedYes

Keywords

  • Conditional least squares estimator
  • Delay parameter
  • Nonlinear time series
  • Nonparametric estimator
  • Stationary time series

ASJC Scopus subject areas

  • Statistics and Probability

Fingerprint Dive into the research topics of 'A diagnostic statistic for functional-coefficient autoregressive models'. Together they form a unique fingerprint.

Cite this