Bayesian Markov Regime-Switching Models for Cointegration

作     者：Kai Cui Wenshan Cui 

作者机构：Department of Statistical Science Duke University Durham USA School of Science and Information Qingdao Agricultural University Qingdao China 

出 版 物：《Applied Mathematics》 (应用数学（英文）)

年 卷 期：2012年第3卷第12期

页      面：1892-1897页

学科分类：0202[经济学-应用经济学] 02[经济学] 

主　　题：Cointegration Regime-Switching Bayesian MCMC 

摘      要：This paper introduces a Bayesian Markov regime-switching model that allows the cointegration relationship between two time series to be switched on and off over time. Unlike classical approaches for testing and modeling cointegration, the Bayesian Markov switching method allows for estimation of the regime-specific model parameters via Markov Chain Monte Carlo and generates more reliable estimation. Inference of regime switching also provides important information for further analysis and decision making.