Bridge Proxy-SVAR: Estimating the Macroeconomic Effects of Shocks Identified at High-Frequency
54 Pages Posted: 28 May 2020
Date Written: April 27, 2020
This paper proposes a novel methodology, the Bridge Proxy-SVAR, which exploits high-frequency information for the identification of the Vector Autoregressive (VAR) models employed in macroeconomic analysis. The methodology is comprised of three steps: (I) identifying the structural shocks of interest in high-frequency systems; (II) aggregating the series of high-frequency shocks at a lower frequency; and (III) using the aggregated series of shocks as a proxy for the corresponding structural shock in lower frequency VARs. We show that the methodology correctly recovers the impact effect of the shocks, both formally and in Monte Carlo experiments. Thus the Bridge Proxy-SVAR can improve causal inference in macroeconomics that typically relies on VARs identified at low-frequency. In an empirical application, we identify uncertainty shocks in the U.S. by imposing weaker restrictions relative to the existing literature and find that they induce mildly recessionary effects.
Keywords: structural vector autoregression, external instrument, high-frequency identification, proxy variable, uncertainty shocks
JEL Classification: C32, C36, E32
Suggested Citation: Suggested Citation