Double Robustness of Local Projections and Some Unpleasant VARithmetic

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Olea, José Luis Montiel, Plagborg-Møller, Mikkel, Qian, Eric, Wolf, Christian K.
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912818993823744
author Olea, José Luis Montiel
Plagborg-Møller, Mikkel
Qian, Eric
Wolf, Christian K.
author_facet Olea, José Luis Montiel
Plagborg-Møller, Mikkel
Qian, Eric
Wolf, Christian K.
contents We consider impulse response inference in a locally misspecified vector autoregression (VAR) model. The conventional local projection (LP) confidence interval has correct coverage even when the misspecification is so large that it can be detected with probability approaching 1. This result follows from a "double robustness" property analogous to that of popular partially linear regression estimators. By contrast, the conventional VAR confidence interval with short-to-moderate lag length can severely undercover for misspecification that is small, difficult to detect statistically, and cannot be ruled out based on economic theory. The VAR confidence interval has robust coverage if, and only if, the lag length is so large that the interval is as wide as the LP interval.
format Preprint
id arxiv_https___arxiv_org_abs_2405_09509
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Double Robustness of Local Projections and Some Unpleasant VARithmetic
Olea, José Luis Montiel
Plagborg-Møller, Mikkel
Qian, Eric
Wolf, Christian K.
Econometrics
We consider impulse response inference in a locally misspecified vector autoregression (VAR) model. The conventional local projection (LP) confidence interval has correct coverage even when the misspecification is so large that it can be detected with probability approaching 1. This result follows from a "double robustness" property analogous to that of popular partially linear regression estimators. By contrast, the conventional VAR confidence interval with short-to-moderate lag length can severely undercover for misspecification that is small, difficult to detect statistically, and cannot be ruled out based on economic theory. The VAR confidence interval has robust coverage if, and only if, the lag length is so large that the interval is as wide as the LP interval.
title Double Robustness of Local Projections and Some Unpleasant VARithmetic
topic Econometrics
url https://arxiv.org/abs/2405.09509