A Bayesian Infinite Hidden Markov Vector Autoregressive Model

Tinbergen Institute Discussion Paper 16-107/III

49 Pages Posted: 8 Dec 2016 Last revised: 22 Oct 2017

See all articles by Didier Nibbering

Didier Nibbering

Erasmus University Rotterdam (EUR) - Department of Econometrics

Richard Paap

Erasmus University Rotterdam (EUR) - Department of Econometrics; Tinbergen Institute; Erasmus Research Institute of Management (ERIM)

Michel van der Wel

Erasmus University Rotterdam; CREATES; ERIM; Tinbergen Institute

Date Written: October 13, 2017

Abstract

We propose a Bayesian infinite hidden Markov model to estimate time-varying parameters in a vector autoregressive model. The Markov structure allows for heterogeneity over time while accounting for state-persistence. By modelling the transition distribution as a Dirichlet process mixture model, parameters can vary over potentially an infinite number of regimes. The Dirichlet process however favours a parsimonious model without imposing restrictions on the parameter space. An empirical application demonstrates the ability of the model to capture both smooth and abrupt parameter changes over time, and a real-time forecasting exercise shows excellent predictive performance even in large dimensional VARs.

Keywords: Time-Varying Parameter Vector Autoregressive Model, Semi-parametric Bayesian Inference, Dirichlet Process Mixture Model, Hidden Markov Chain, Monetary Policy Analysis, Real-time Forecasting

JEL Classification: C11, C14, C32, C51, C54

Suggested Citation

Nibbering, Didier and Paap, Richard and van der Wel, Michel, A Bayesian Infinite Hidden Markov Vector Autoregressive Model (October 13, 2017). Tinbergen Institute Discussion Paper 16-107/III. Available at SSRN: https://ssrn.com/abstract=2882463 or http://dx.doi.org/10.2139/ssrn.2882463

Didier Nibbering (Contact Author)

Erasmus University Rotterdam (EUR) - Department of Econometrics ( email )

P.O. Box 1738
3000 DR Rotterdam
Netherlands

Richard Paap

Erasmus University Rotterdam (EUR) - Department of Econometrics ( email )

P.O. Box 1738
3000 DR Rotterdam
Netherlands

Tinbergen Institute ( email )

P.O. Box 1738
3000 DR Rotterdam
Netherlands

Erasmus Research Institute of Management (ERIM) ( email )

P.O. Box 1738
3000 DR Rotterdam
Netherlands

Michel Van der Wel

Erasmus University Rotterdam ( email )

Burg. Oudlaan 50
Rotterdam, NL 3062 PA
Netherlands

CREATES ( email )

School of Economics and Management
Building 1322, Bartholins Alle 10
DK-8000 Aarhus C
Denmark

ERIM ( email )

P.O. Box 1738
3000 DR Rotterdam
Netherlands

Tinbergen Institute ( email )

Burg. Oudlaan 50
Rotterdam, 3062 PA
Netherlands

Register to save articles to
your library

Register

Paper statistics

Downloads
82
Abstract Views
507
rank
298,943
PlumX Metrics