Testing a linear ARMA model against threshold-ARMA models: A Bayesian approach

ABSTRACT We introduce a Bayesian approach to test linear autoregressive moving-average (ARMA) models against threshold autoregressive moving-average (TARMA) models. First, the marginal posterior densities of all parameters, including the threshold and delay, of a TARMA model are obtained by using Gibbs sampler with Metropolis–Hastings algorithm. Second, reversible-jump Markov chain Monte Carlo (RJMCMC) method is adopted to calculate the posterior probabilities for ARMA and TARMA models: Posterior evidence in favor of TARMA models indicates threshold nonlinearity. Finally, based on RJMCMC scheme and Akaike information criterion (AIC) or Bayesian information criterion (BIC), the procedure for modeling TARMA models is exploited. Simulation experiments and a real data example show that our method works well for distinguishing an ARMA from a TARMA model and for building TARMA models.

[1]  Mohamed A. Ismail,et al.  Bayesian inference for threshold moving average models , 2003 .

[2]  G. Casella,et al.  Explaining the Gibbs Sampler , 1992 .

[3]  Cathy W. S. Chen,et al.  Subset threshold autoregression , 2003 .

[4]  John Geweke,et al.  BAYESIAN THRESHOLD AUTOREGRESSIVE MODELS FOR NONLINEAR TIME SERIES , 1993 .

[5]  Cathy W. S. Chen,et al.  BAYESIAN INFERENCE OF THRESHOLD AUTOREGRESSIVE MODELS , 1995 .

[6]  Cathy W. S. Chen.  Subset selection of autoregressive time series models , 1999 .

[7]  N. Metropolis,et al.  Equation of State Calculations by Fast Computing Machines , 1953, Resonance.

[8]  E. George,et al.  Journal of the American Statistical Association is currently published by American Statistical Association. , 2007 .

[9]  P. Morettin.  BAYESIAN ANALYSIS OF THRESHOLD AUTOREGRESSIVE MOVING AVERAGE MODELS , 2002 .

[10]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Bayesian Analysis of Two-Regime Threshold Autoregressive Moving Average Model with Exogenous Inputs , 2012 .

[12]  Ioannis D. Vrontos,et al.  Full Bayesian Inference for GARCH and EGARCH Models , 2000 .

[13]  Paul Waltman,et al.  A Threshold Model , 1974 .

[14]  Bin Yu,et al.  Looking at Markov samplers through cusum path plots: a simple diagnostic idea , 1998, Stat. Comput..

[15]  H. Tong,et al.  Threshold Autoregression, Limit Cycles and Cyclical Data , 1980 .

[16]  Jiazhu Pan,et al.  A Bayesian nonlinearity test for threshold moving average models , 2010 .

[17]  H. Akaike.  A new look at the statistical model identification , 1974 .

[18]  R. McCulloch,et al.  Bayesian Inference and Prediction for Mean and Variance Shifts in Autoregressive Time Series , 1993 .

[19]  Eric R. Ziegel,et al.  Analysis of Financial Time Series , 2002, Technometrics.

[20]  S. Chib,et al.  Understanding the Metropolis-Hastings Algorithm , 1995 .

[21]  Cathy W. S. Chen,et al.  Bayesian Model Selection for Heteroskedastic Models , 2008 .

[22]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[23]  H. Tong.  Non-linear time series. A dynamical system approach , 1990 .

[24]  Cathy W. S. Chen,et al.  Best subset selection of autoregressive models with exogenous variables and generalized autoregressive conditional heteroscedasticity errors , 2006 .

[25]  刘金山,et al.  Bayesian analysis of two-regime threshold autoregressive-moving average models with exogenous inputs , 2012 .

[26]  Cathy W. S. Chen,et al.  A Bayesian threshold nonlinearity test for financial time series , 2005 .

[27]  Howell Tong,et al.  TESTING FOR A LINEAR MA MODEL AGAINST THRESHOLD MA MODELS , 2005 .

[28]  Cathy W. S. Chen,et al.  Bayesian subset selection for threshold autoregressive moving-average models , 2011, Comput. Stat..

[29]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .