BAYESIAN THRESHOLD AUTOREGRESSIVE MODELS FOR NONLINEAR TIME SERIES

John Geweke, Nobuhiko Terui

Research output: Contribution to journalArticlepeer-review

79 Citations (Scopus)

Abstract

Abstract. This paper provides a Bayesian approach to statistical inference in the threshold autoregressive model for time series. The exact posterior distribution of the delay and threshold parameters is derived, as is the multi‐step‐ahead predictive density. The proposed methods are applied to the Wolfe's sunspot and Canadian lynx data sets.

Original languageEnglish
Pages (from-to)441-454
Number of pages14
JournalJournal of Time Series Analysis
Volume14
Issue number5
DOIs
Publication statusPublished - 1993 Sep
Externally publishedYes

Keywords

  • Canadian lynx
  • Monte Carlo integration
  • Nonlinear time series
  • Wolfe's sunspot
  • regime change prediction
  • threshold model

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Fingerprint Dive into the research topics of 'BAYESIAN THRESHOLD AUTOREGRESSIVE MODELS FOR NONLINEAR TIME SERIES'. Together they form a unique fingerprint.

Cite this