Estimating (Markov-Switching) VAR Models Without Gibbs Sampling: A Sequential Monte Carlo Approach

56 Pages Posted: 22 Dec 2015

See all articles by Mark Bognanni

Mark Bognanni

Federal Reserve Banks - Federal Reserve Bank of Cleveland

Edward Herbst

Board of Governors of the Federal Reserve System

Multiple version iconThere are 2 versions of this paper

Date Written: 2015-12-18

Abstract

Vector autoregressions with Markov-switching parameters (MS-VARs) fit the data better than do their constant-parameter predecessors. However, Bayesian inference for MS-VARs with existing algorithms remains challenging. For our first contribution, we show that Sequential Monte Carlo (SMC) estimators accurately estimate Bayesian MS-VAR posteriors. Relative to multi-step, model-specific MCMC routines, SMC has the advantages of generality, parallelizability, and freedom from reliance on particular analytical relationships between prior and likelihood. For our second contribution, we use SMC's flexibility to demonstrate that the choice of prior drives the key empirical finding of Sims, Waggoner, and Zha (2008) as much as does the data.

Keywords: Bayesian Analysis, Regime-Switching Models, Sequential Monte Carlo, Vector Autoregressions

JEL Classification: C11, C18, C32, C52, E3, E4, E5

Suggested Citation

Bognanni, Mark and Herbst, Edward, Estimating (Markov-Switching) VAR Models Without Gibbs Sampling: A Sequential Monte Carlo Approach (2015-12-18). FEDS Working Paper No. 2015-116. Available at SSRN: https://ssrn.com/abstract=2705735 or http://dx.doi.org/10.17016/FEDS.2015.116

Mark Bognanni (Contact Author)

Federal Reserve Banks - Federal Reserve Bank of Cleveland ( email )

East 6th & Superior
Cleveland, OH 44101-1387
United States

Edward Herbst

Board of Governors of the Federal Reserve System ( email )

20th Street and Constitution Avenue NW
Washington, DC 20551
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
53
Abstract Views
272
rank
200,725
PlumX Metrics