Statistical Inference in Markov Switching Vector Error Correction Model Using a Markov Chain Monte Carlo Method

摘要

This paper introduces statistical inference in a Markov switching vector error correction model using a Markov chain Monte Carlo method. The proposed model allows for regime shifts in the deterministic terms, the lag terms, the adjustment terms and the variance-covariance matrix. The proposed method allows for estimation of the cointegrating vector within a nonlinear framework through a collapsed Gibbs sampling. We apply the proposed model to U.S. term structure of interest rates.