Inference for Impulse Responses under Model Uncertainty

TL;DR

This paper introduces a new data-driven method called WIMP for constructing confidence intervals for impulse responses in macroeconomic VAR models that accounts for cointegration rank uncertainty, improving robustness and accuracy.

Contribution

The paper develops the WIMP method to handle model uncertainty in impulse response analysis, outperforming existing methods in simulations and providing more reliable macroeconomic inferences.

Findings

WIMP produces more robust confidence intervals under rank uncertainty.

WIMP outperforms existing methods in simulation studies.

Re-assessment of fiscal policy shocks shows importance of accounting for rank uncertainty.

Abstract

In many macroeconomic applications, confidence intervals for impulse responses are constructed by estimating VAR models in levels - ignoring cointegration rank uncertainty. We investigate the consequences of ignoring this uncertainty. We adapt several methods for handling model uncertainty and highlight their shortcomings. We propose a new method - Weighted-Inference-by-Model-Plausibility (WIMP) - that takes rank uncertainty into account in a data-driven way. In simulations the WIMP outperforms all other methods considered, delivering intervals that are robust to rank uncertainty, yet not overly conservative. We also study potential ramifications of rank uncertainty on applied macroeconomic analysis by re-assessing the effects of fiscal policy shocks.