A Bayesian detection of structural changes in autoregressive time series models

This study investigates a Bayesian detection of a change in any parameter, or in any collection of parameters of the autoregressive time series model of known order p. An unconditional Bayesian test based on the highest posterior density credible sets is determined. Using the Gibbs sampler algorithm, some simulated results are given to approximate the posterior densities of the change point and other parameters of the model. The performance of our proposed method has been investigated on simulated and real data sets.