Wild inference for wild SVARs with application to heteroscedasticity-based IV

TL;DR

This paper introduces a dependent wild bootstrap method for inference on impulse response functions in nonstationary, heteroscedastic SVARs, avoiding pretesting and enabling analysis with local projections in complex data.

Contribution

It proposes a novel bootstrap procedure for IRF inference in nonstationary, heteroscedastic SVARs that eliminates the need for pretesting and improves smoothing of estimates.

Findings

Validates the method with DSGE simulations

Applies to US monetary policy using FOMC data

Demonstrates robustness in heteroscedastic, nonstationary settings

Abstract

Structural vector autoregressions are used to compute impulse response functions (IRF) for persistent data. Existing multiple-parameter inference requires cumbersome pretesting for unit roots, cointegration, and trends with subsequent stationarization. To avoid pretesting, we propose a novel \emph{dependent wild bootstrap} procedure for simultaneous inference on IRF using local projections (LP) estimated in levels in possibly \emph{nonstationary} and \emph{heteroscedastic} SVARs. The bootstrap also allows efficient smoothing of LP estimates. We study IRF to US monetary policy identified using FOMC meetings count as an instrument for heteroscedasticity of monetary shocks. We validate our method using DSGE model simulations and alternative SVAR methods.