Methods for inference in large multiple-equation Markov-switching models

Christopher A. Sims, Daniel F. Waggoner, Tao Zha

Research output: Contribution to journalArticle

91 Scopus citations

Abstract

Inference for multiple-equation Markov-chain models raises a number of difficulties that are unlikely to appear in smaller models. Our framework allows for many regimes in the transition matrix, without letting the number of free parameters grow as the square as the number of regimes, but also without losing a convenient form for the posterior distribution. Calculation of marginal data densities is difficult in these high-dimensional models. This paper gives methods to overcome these difficulties, and explains why existing methods are unreliable. It makes suggestions for maximizing posterior density and initiating MCMC simulations that provide robustness against the complex likelihood shape.

Original languageEnglish (US)
Pages (from-to)255-274
Number of pages20
JournalJournal of Econometrics
Volume146
Issue number2
DOIs
StatePublished - Oct 1 2008

All Science Journal Classification (ASJC) codes

  • Economics and Econometrics

Keywords

  • Composite Markov process
  • Density overlap
  • Incremental and discontinuous changes
  • Integrated-out likelihood
  • New MHM

Fingerprint Dive into the research topics of 'Methods for inference in large multiple-equation Markov-switching models'. Together they form a unique fingerprint.

  • Cite this