VAR Forecasting Using Bayesian Variable Selection

33 Pages Posted: 5 Mar 2010 Last revised: 23 Sep 2014

See all articles by Dimitris Korobilis

Dimitris Korobilis

University of Glasgow - Adam Smith Business School

Date Written: December 1, 2009

Abstract

This paper develops methods for automatic selection of variables in Bayesian vector autoregressions (VARs) using the Gibbs sampler. In particular, I provide computationally efficient algorithms for stochastic variable selection in generic linear and nonlinear models, as well as models of large dimensions. The performance of the proposed variable selection method is assessed in forecasting three major macroeconomic time series of the UK economy. Data-based restrictions of VAR coefficients can help improve upon their unrestricted counterparts in forecasting, and in many cases they compare favorably to shrinkage estimators.

Keywords: Forecasting, variable selection, time-varying parameters, Bayesian vector autoregression

JEL Classification: C11, C32, C52, C53, E37

Suggested Citation

Korobilis, Dimitris, VAR Forecasting Using Bayesian Variable Selection (December 1, 2009). Available at SSRN: https://ssrn.com/abstract=1564378 or http://dx.doi.org/10.2139/ssrn.1564378

Dimitris Korobilis (Contact Author)

University of Glasgow - Adam Smith Business School ( email )

40 University Avenue
Gilbert Scott Building
Glasgow, Scotland G12 8QQ
United Kingdom

HOME PAGE: http://https://sites.google.com/site/dimitriskorobilis/

Register to save articles to
your library

Register

Paper statistics

Downloads
200
Abstract Views
729
rank
155,453
PlumX Metrics