A Bayesian nonlinearity test for threshold moving average models

Qing Xia, Jiazhu Pan, Zhiqiang Zhang, Jinshan Liu

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

We propose a Bayesian test for nonlinearity of threshold moving average (TMA) models. First, we obtain the marginal posterior densities of all parameters, including the threshold and delay, of the TMA model using Gibbs sampler with the Metropolis-Hastings algorithm. And then, we adopt reversible-jump Markov chain Monte Carlo methods to calculate the posterior probabilities for MA and TMA models. Posterior evidence in favour of the TMA model indicates threshold nonlinearity. Simulation experiments and a real example show that our method works very well in distinguishing MA and TMA models.
LanguageEnglish
Pages329-336
Number of pages8
JournalJournal of Time Series Analysis
Volume31
Issue number5
DOIs
Publication statusPublished - 1 Sep 2010

Fingerprint

Moving Average Model
Threshold Model
Nonlinearity
Reversible Jump Markov Chain Monte Carlo
Metropolis-Hastings Algorithm
Gibbs Sampler
Markov Chain Monte Carlo Methods
Posterior Probability
Markov processes
Simulation Experiment
Monte Carlo methods
Nonlinearity test
Moving average
Calculate
Experiments

Keywords

  • MA models
  • TMA models
  • RJMCMC methods
  • Metropolis-Hastings
  • Gibbs sampler
  • MA models
  • Bayesian inference

Cite this

Xia, Qing ; Pan, Jiazhu ; Zhang, Zhiqiang ; Liu, Jinshan. / A Bayesian nonlinearity test for threshold moving average models. In: Journal of Time Series Analysis. 2010 ; Vol. 31, No. 5. pp. 329-336.
@article{87a31e7e82514ecba83d94220f594143,
title = "A Bayesian nonlinearity test for threshold moving average models",
abstract = "We propose a Bayesian test for nonlinearity of threshold moving average (TMA) models. First, we obtain the marginal posterior densities of all parameters, including the threshold and delay, of the TMA model using Gibbs sampler with the Metropolis-Hastings algorithm. And then, we adopt reversible-jump Markov chain Monte Carlo methods to calculate the posterior probabilities for MA and TMA models. Posterior evidence in favour of the TMA model indicates threshold nonlinearity. Simulation experiments and a real example show that our method works very well in distinguishing MA and TMA models.",
keywords = "MA models, TMA models , RJMCMC methods, Metropolis-Hastings, Gibbs sampler, MA models, Bayesian inference",
author = "Qing Xia and Jiazhu Pan and Zhiqiang Zhang and Jinshan Liu",
year = "2010",
month = "9",
day = "1",
doi = "10.1111/j.1467-9892.2010.00667.x",
language = "English",
volume = "31",
pages = "329--336",
journal = "Journal of Time Series Analysis",
issn = "0143-9782",
number = "5",

}

A Bayesian nonlinearity test for threshold moving average models. / Xia, Qing; Pan, Jiazhu; Zhang, Zhiqiang; Liu, Jinshan.

In: Journal of Time Series Analysis, Vol. 31, No. 5, 01.09.2010, p. 329-336.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Bayesian nonlinearity test for threshold moving average models

AU - Xia, Qing

AU - Pan, Jiazhu

AU - Zhang, Zhiqiang

AU - Liu, Jinshan

PY - 2010/9/1

Y1 - 2010/9/1

N2 - We propose a Bayesian test for nonlinearity of threshold moving average (TMA) models. First, we obtain the marginal posterior densities of all parameters, including the threshold and delay, of the TMA model using Gibbs sampler with the Metropolis-Hastings algorithm. And then, we adopt reversible-jump Markov chain Monte Carlo methods to calculate the posterior probabilities for MA and TMA models. Posterior evidence in favour of the TMA model indicates threshold nonlinearity. Simulation experiments and a real example show that our method works very well in distinguishing MA and TMA models.

AB - We propose a Bayesian test for nonlinearity of threshold moving average (TMA) models. First, we obtain the marginal posterior densities of all parameters, including the threshold and delay, of the TMA model using Gibbs sampler with the Metropolis-Hastings algorithm. And then, we adopt reversible-jump Markov chain Monte Carlo methods to calculate the posterior probabilities for MA and TMA models. Posterior evidence in favour of the TMA model indicates threshold nonlinearity. Simulation experiments and a real example show that our method works very well in distinguishing MA and TMA models.

KW - MA models

KW - TMA models

KW - RJMCMC methods

KW - Metropolis-Hastings

KW - Gibbs sampler

KW - MA models

KW - Bayesian inference

UR - http://www.scopus.com/inward/record.url?scp=77955757177&partnerID=8YFLogxK

U2 - 10.1111/j.1467-9892.2010.00667.x

DO - 10.1111/j.1467-9892.2010.00667.x

M3 - Article

VL - 31

SP - 329

EP - 336

JO - Journal of Time Series Analysis

T2 - Journal of Time Series Analysis

JF - Journal of Time Series Analysis

SN - 0143-9782

IS - 5

ER -