Forecasting with medium and large Bayesian VARs

Research output: Contribution to journalArticle

86 Citations (Scopus)

Abstract

This paper is motivated by the recent interest in the use of Bayesian VARs for forecasting, even in cases where the number of dependent variables is large. In such cases factor methods have been traditionally used, but recent work using a particular prior suggests that Bayesian VAR methods can forecast better. In this paper, we consider a range of alternative priors which have been used with small VARs, discuss the issues which arise when they are used with medium and large VARs and examine their forecast performance using a US macroeconomic dataset containing 168 variables. We find that Bayesian VARs do tend to forecast better than factor methods and provide an extensive comparison of the strengths and weaknesses of various approaches. Typically, we find that the simple Minnesota prior forecasts well in medium and large VARs, which makes this prior attractive relative to computationally more demanding alternatives. Our empirical results show the importance of using forecast metrics based on the entire predictive density, instead of relying solely on those based on point forecasts
LanguageEnglish
Number of pages27
JournalJournal of Applied Econometrics
Early online date17 Oct 2011
DOIs
Publication statusPublished - 2012

Fingerprint

macroeconomics
performance
Bayesian VAR
Factors
Empirical results
Macroeconomics
Point forecasts
Predictive density
Forecast performance

Keywords

  • Bayesian VARs
  • forecasting
  • priors
  • small VARs
  • computational issues
  • conditionally conjugate
  • VAR forecasting
  • empirical study

Cite this

@article{e0fda3d1b6b74deba7c2da5a737edd66,
title = "Forecasting with medium and large Bayesian VARs",
abstract = "This paper is motivated by the recent interest in the use of Bayesian VARs for forecasting, even in cases where the number of dependent variables is large. In such cases factor methods have been traditionally used, but recent work using a particular prior suggests that Bayesian VAR methods can forecast better. In this paper, we consider a range of alternative priors which have been used with small VARs, discuss the issues which arise when they are used with medium and large VARs and examine their forecast performance using a US macroeconomic dataset containing 168 variables. We find that Bayesian VARs do tend to forecast better than factor methods and provide an extensive comparison of the strengths and weaknesses of various approaches. Typically, we find that the simple Minnesota prior forecasts well in medium and large VARs, which makes this prior attractive relative to computationally more demanding alternatives. Our empirical results show the importance of using forecast metrics based on the entire predictive density, instead of relying solely on those based on point forecasts",
keywords = "Bayesian VARs, forecasting, priors, small VARs, computational issues, conditionally conjugate, VAR forecasting, empirical study",
author = "Gary Koop",
year = "2012",
doi = "10.1002/jae.1270",
language = "English",
journal = "Journal of Applied Econometrics",
issn = "0883-7252",

}

Forecasting with medium and large Bayesian VARs. / Koop, Gary.

In: Journal of Applied Econometrics, 2012.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Forecasting with medium and large Bayesian VARs

AU - Koop, Gary

PY - 2012

Y1 - 2012

N2 - This paper is motivated by the recent interest in the use of Bayesian VARs for forecasting, even in cases where the number of dependent variables is large. In such cases factor methods have been traditionally used, but recent work using a particular prior suggests that Bayesian VAR methods can forecast better. In this paper, we consider a range of alternative priors which have been used with small VARs, discuss the issues which arise when they are used with medium and large VARs and examine their forecast performance using a US macroeconomic dataset containing 168 variables. We find that Bayesian VARs do tend to forecast better than factor methods and provide an extensive comparison of the strengths and weaknesses of various approaches. Typically, we find that the simple Minnesota prior forecasts well in medium and large VARs, which makes this prior attractive relative to computationally more demanding alternatives. Our empirical results show the importance of using forecast metrics based on the entire predictive density, instead of relying solely on those based on point forecasts

AB - This paper is motivated by the recent interest in the use of Bayesian VARs for forecasting, even in cases where the number of dependent variables is large. In such cases factor methods have been traditionally used, but recent work using a particular prior suggests that Bayesian VAR methods can forecast better. In this paper, we consider a range of alternative priors which have been used with small VARs, discuss the issues which arise when they are used with medium and large VARs and examine their forecast performance using a US macroeconomic dataset containing 168 variables. We find that Bayesian VARs do tend to forecast better than factor methods and provide an extensive comparison of the strengths and weaknesses of various approaches. Typically, we find that the simple Minnesota prior forecasts well in medium and large VARs, which makes this prior attractive relative to computationally more demanding alternatives. Our empirical results show the importance of using forecast metrics based on the entire predictive density, instead of relying solely on those based on point forecasts

KW - Bayesian VARs

KW - forecasting

KW - priors

KW - small VARs

KW - computational issues

KW - conditionally conjugate

KW - VAR forecasting

KW - empirical study

UR - http://www.scopus.com/inward/record.url?scp=84874021974&partnerID=8YFLogxK

U2 - 10.1002/jae.1270

DO - 10.1002/jae.1270

M3 - Article

JO - Journal of Applied Econometrics

T2 - Journal of Applied Econometrics

JF - Journal of Applied Econometrics

SN - 0883-7252

ER -