Granger-Causal-Priority and Choice of Variables in Vector Autoregressions

56 Pages Posted: 8 Nov 2013

See all articles by Marek Jarocinski

Marek Jarocinski

European Central Bank (ECB)

Bartosz Maćkowiak

European Central Bank (ECB)

Multiple version iconThere are 2 versions of this paper

Date Written: October 15, 2013

Abstract

A researcher is interested in a set of variables that he wants to model with a vector auto-regression and he has a dataset with more variables. Which variables from the dataset to include in the VAR, in addition to the variables of interest? This question arises in many applications of VARs, in prediction and impulse response analysis. We develop a Bayesian methodology to answer this question. We rely on the idea of Granger-causal-priority, related to the well-known concept of Granger-non-causality. The methodology is simple to use, because we provide closed-form expressions for the relevant posterior probabilities. Applying the methodology to the case when the variables of interest are output, the price level, and the short-term interest rate, we find remarkably similar results for the United States and the euro area.

Keywords: Vector autoregression, structural vector autoregression, Granger-causalpriority, Granger-noncausality, Bayesian model choice

JEL Classification: C32, C52, E32

Suggested Citation

Jarocinski, Marek and Maćkowiak, Bartosz, Granger-Causal-Priority and Choice of Variables in Vector Autoregressions (October 15, 2013). ECB Working Paper No. 1600. Available at SSRN: https://ssrn.com/abstract=2340402

Marek Jarocinski (Contact Author)

European Central Bank (ECB) ( email )

Sonnemannstrasse 22
Frankfurt am Main, 60314
Germany
+49 69 1344 6414 (Phone)

Bartosz Maćkowiak

European Central Bank (ECB) ( email )

Sonnemannstrasse 22
Frankfurt am Main, 60314
Germany

Register to save articles to
your library

Register

Paper statistics

Downloads
53
Abstract Views
775
rank
384,354
PlumX Metrics