Instrumental Variable Identification of Dynamic Variance Decompositions

Mikkel Plagborg-Møller; Christian K. Wolf

doi:10.1086/720141

Instrumental Variable Identification of Dynamic Variance Decompositions

Mikkel Plagborg-Møller, Christian K. Wolf

Princeton University

Research output: Contribution to journal › Article › peer-review

Abstract

Macroeconomists increasingly use external sources of exogenous variation for causal inference. However, unless such external instruments (proxies) capture the underlying shock without measurement error, existing methods are silent on the importance of that shock for macroeconomic fluctuations. We show that, in a general moving-average model with external instruments, variance decompositions for the instrumented shock are interval-identified, with informative bounds. Various additional restrictions guarantee point identification of both variance and historical decompositions. Unlike structural vector auto-regression analysis, our methods do not require invertibility. Applied to US data, they give a tight upper bound on the importance of monetary shocks for inflation dynamics.

Original language	English (US)
Pages (from-to)	2164-2202
Number of pages	39
Journal	Journal of Political Economy
Volume	130
Issue number	8
DOIs	https://doi.org/10.1086/720141
State	Published - Aug 2022

All Science Journal Classification (ASJC) codes

Economics and Econometrics

Access to Document

10.1086/720141

Cite this

@article{4a486afed88b4caaa0a647207f8e80f3,

title = "Instrumental Variable Identification of Dynamic Variance Decompositions",

abstract = "Macroeconomists increasingly use external sources of exogenous variation for causal inference. However, unless such external instruments (proxies) capture the underlying shock without measurement error, existing methods are silent on the importance of that shock for macroeconomic fluctuations. We show that, in a general moving-average model with external instruments, variance decompositions for the instrumented shock are interval-identified, with informative bounds. Various additional restrictions guarantee point identification of both variance and historical decompositions. Unlike structural vector auto-regression analysis, our methods do not require invertibility. Applied to US data, they give a tight upper bound on the importance of monetary shocks for inflation dynamics.",

author = "Mikkel Plagborg-M{\o}ller and Wolf, {Christian K.}",

year = "2022",

month = aug,

doi = "10.1086/720141",

language = "English (US)",

volume = "130",

pages = "2164--2202",

journal = "Journal of Political Economy",

issn = "0022-3808",

publisher = "University of Chicago",

number = "8",

}

TY - JOUR

T1 - Instrumental Variable Identification of Dynamic Variance Decompositions

AU - Plagborg-Møller, Mikkel

AU - Wolf, Christian K.

PY - 2022/8

Y1 - 2022/8

N2 - Macroeconomists increasingly use external sources of exogenous variation for causal inference. However, unless such external instruments (proxies) capture the underlying shock without measurement error, existing methods are silent on the importance of that shock for macroeconomic fluctuations. We show that, in a general moving-average model with external instruments, variance decompositions for the instrumented shock are interval-identified, with informative bounds. Various additional restrictions guarantee point identification of both variance and historical decompositions. Unlike structural vector auto-regression analysis, our methods do not require invertibility. Applied to US data, they give a tight upper bound on the importance of monetary shocks for inflation dynamics.

AB - Macroeconomists increasingly use external sources of exogenous variation for causal inference. However, unless such external instruments (proxies) capture the underlying shock without measurement error, existing methods are silent on the importance of that shock for macroeconomic fluctuations. We show that, in a general moving-average model with external instruments, variance decompositions for the instrumented shock are interval-identified, with informative bounds. Various additional restrictions guarantee point identification of both variance and historical decompositions. Unlike structural vector auto-regression analysis, our methods do not require invertibility. Applied to US data, they give a tight upper bound on the importance of monetary shocks for inflation dynamics.

UR - http://www.scopus.com/inward/record.url?scp=85131551757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85131551757&partnerID=8YFLogxK

U2 - 10.1086/720141

DO - 10.1086/720141

M3 - Article

SN - 0022-3808

VL - 130

SP - 2164

EP - 2202

JO - Journal of Political Economy

JF - Journal of Political Economy

IS - 8

ER -

Instrumental Variable Identification of Dynamic Variance Decompositions

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this