Bayesian Model Uncertainty in Smooth Transition Autoregressions

19 Pages Posted: 21 Feb 2006  

Hedibert F. Lopes

University of Chicago - Booth School of Business

Esther Salazar

Duke University; Universidade Federal do Rio de Janeiro (UFRJ)

Abstract

In this paper, we propose a fully Bayesian approach to the special class of nonlinear time-series models called the logistic smooth transition autoregressive (LSTAR) model. Initially, a Gibbs sampler is proposed for the LSTAR where the lag length, k, is kept fixed. Then, uncertainty about k is taken into account and a novel reversible jump Markov Chain Monte Carlo (RJMCMC) algorithm is proposed. We compared our RJMCMC algorithm with well-known information criteria, such as the Akaike information criteria, the Bayesian information criteria (BIC) and the deviance information criteria. Our methodology is extensively studied against simulated and real-time series.

Keywords: Markov Chain Monte Carlo, nonlinear time-series model, model selection, reversible jump MCMC; deviance information criterion

Suggested Citation

Lopes, Hedibert F. and Salazar, Esther, Bayesian Model Uncertainty in Smooth Transition Autoregressions. Journal of Time Series Analysis, Vol. 27, No. 1, pp. 99-117, January 2006. Available at SSRN: https://ssrn.com/abstract=875065 or http://dx.doi.org/10.1111/j.1467-9892.2005.00455.x

Hedibert F. Lopes (Contact Author)

University of Chicago - Booth School of Business ( email )

5807 S. Woodlawn Avenue
Chicago, IL 60637
United States

Esther Salazar

Duke University ( email )

Durham, NC 27708
United States

HOME PAGE: http://www.duke.edu/~es145

Universidade Federal do Rio de Janeiro (UFRJ) ( email )

Rua General Canabarro, 706
terreo - Bairro Maracana
20271-201 - Rio de Janeiro, RJ, 23890000
Brazil

Paper statistics

Downloads
9
Abstract Views
1,020