Home About Latest Browse RSS Advanced Search

Board of Governors of the Federal Reserve System (US)
Finance and Economics Discussion Series
Estimating (Markov-Switching) VAR Models without Gibbs Sampling: A Sequential Monte Carlo Approach
Mark Bognanni
Edward Herbst
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.


Download Full text
Download DOI
Cite this item
Mark Bognanni & Edward Herbst, Estimating (Markov-Switching) VAR Models without Gibbs Sampling: A Sequential Monte Carlo Approach, Board of Governors of the Federal Reserve System (US), Finance and Economics Discussion Series 2015-116, 18 Dec 2015.
More from this series
JEL Classification:
Subject headings:
Keywords: Bayesian Analysis; Regime-Switching Models; Sequential Monte Carlo; Vector Autoregressions
DOI: 10.17016/FEDS.2015.116
For corrections, contact Ryan Wolfslayer ()
Fed-in-Print is the central catalog of publications within the Federal Reserve System. It is managed and hosted by the Economic Research Division, Federal Reserve Bank of St. Louis.

Privacy Legal