Working Paper

Asymptotically Valid Bootstrap Inference for Proxy SVARs

Abstract: Proxy structural vector autoregressions identify structural shocks in vector autoregressions with external variables that are correlated with the structural shocks of interest but uncorrelated with all other structural shocks. We provide asymptotic theory for this identification approach under mild ?-mixing conditions that cover a large class of uncorrelated, but possibly dependent innovation processes, including conditional heteroskedasticity. We prove consistency of a residual-based moving block bootstrap for inference on statistics such as impulse response functions and forecast error variance decompositions. Wild bootstraps are proven to be generally invalid for these statistics and their coverage rates can be badly and persistently mis-sized.

Keywords: Wild Bootstrap; Mixing; Proxy Variables; Residual-Based Moving Block Bootstrap; Structural Vector Autoregression; External Instruments;

JEL Classification: C30; C32;

https://doi.org/10.26509/frbc-wp-201908

Bibliographic Information

Provider: Federal Reserve Bank of Cleveland

Part of Series: Working Papers

Publication Date: 2019-05-03

Number: 19-08