Secret Ingredients? The Informational Contribution of Factors to Standard VAR Analysis

39 Pages Posted: 17 Sep 2010

Date Written: August 27, 2010

Abstract

Recent work with Factor-Augmented Vector Autoregression (FAVAR) suggests that standard VAR analysis can be improved by incorporating the information in a large number of macroeconomic time series. I examine what new information FAVAR factors contribute. Using a sparse modification to principal components, I find that 1) extracted factors (and their impulse responses) have a natural economic interpretation and 2) a particular small-scale VAR specification closely reproduces the results of standard FAVARs on US data. My results suggest that three leading economic indicators - private payroll employment, the NAPM purchasing managers’ index, and housing starts - substantially capture the extra information introduced by FAVAR.

Keywords: Factor-Augmented Vector Autoregression, VAR, FAVAR, Monetary Policy, Principal Components

JEL Classification: E31, E4, E52, C32

Suggested Citation

Chahrour, Ryan A., Secret Ingredients? The Informational Contribution of Factors to Standard VAR Analysis (August 27, 2010). Available at SSRN: https://ssrn.com/abstract=1499884 or http://dx.doi.org/10.2139/ssrn.1499884

Ryan A. Chahrour (Contact Author)

Cornell University ( email )

Ithaca, NY 14853
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
47
Abstract Views
461
PlumX Metrics