A Bayesian analysis of a change in the parameters of autoregressive time series |
| |
Authors: | Abdeldjalil Slama Hafida Saggou |
| |
Institution: | 1. Department of Mathematics and Computer Science, University of Adrar, Adrar, Algeria;2. Department of Probability and Statistics, USTHB, Algiers, Algeria;3. Department of Probability and Statistics, USTHB, Algiers, Algeria |
| |
Abstract: | In this article, we consider a Bayesian analysis of a possible change in the parameters of autoregressive time series of known order p, AR(p). An unconditional Bayesian test based on highest posterior density (HPD) credible sets is determined. The test is useful to detect a change in any one of the parameters separately. Using the Gibbs sampler algorithm, we approximate the posterior densities of the change point and other parameters to calculate the p-values that define our test. |
| |
Keywords: | Autoregressive model Bayesian analysis Change point Gibbs sampler HPD credible set p-Values |
|
|