Working Paper: CEPR ID: DP15529
Authors: Mario Forni; Luca Gambetti; Marco Lippi; Luca Sala
Abstract: Structural VAR models produce results that can vary dramatically with thechoice of variables, because information is deficient and/or contaminated bymeasured errors. We propose a novel procedure, the Common ComponentSVAR (CC-SVAR), which solves both problems. First, the common componentsof the variables of interest are estimated using High-DimensionalFactor techniques. Second, SVAR analysis is performed using such components.The key feature is that number of common components is largerthan the number of shocks, so that the SVAR is singular. Consistency resultsfor singular VARs are provided. We apply our procedure to monetarypolicy shocks. Our finding is that, with the CC-SVAR, results are robustand SVAR puzzles disappear.
Keywords: Structural VAR models; Structural factor models; Nonfundamentalness
JEL Codes: C32; E32
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| CCSVAR approach (C10) | consistent estimates of structural shocks (C51) |
| traditional SVAR models (C32) | unreliable estimates (C51) |
| CCSVAR methodology (C10) | disappearance of puzzles (Y70) |
| contractionary monetary policy shocks (E39) | reduce prices (D49) |
| CCSVAR approach (C10) | robust impulse response functions (IRFs) (C22) |