Testing for weak instruments in Linear Iv regression

James H. Stock; Motohiro Yogo

doi:10.1017/CBO9780511614491.006

Testing for weak instruments in Linear Iv regression

James H. Stock, Motohiro Yogo

Research output: Chapter in Book/Report/Conference proceeding › Chapter

3871 Scopus citations

Abstract

Weak instruments can produce biased IV estimators and hypothesis tests with large size distortions. But what, precisely, are weak instruments, and how does one detect them in practice? This paper proposes quantitative definitions of weak instruments based on the maximum IV estimator bias, or the maximumWald test size distortion, when there are multiple endogenous regressors.We tabulate critical values that enable using the first-stage F-statistic (or, when there are multiple endogenous regressors, the Cragg–Donald [1993] statistic) to test whether the given instruments are weak.

Original language	English (US)
Title of host publication	Identification and Inference for Econometric Models
Subtitle of host publication	Essays in Honor of Thomas Rothenberg
Publisher	Cambridge University Press
Pages	80-108
Number of pages	29
ISBN (Electronic)	9780511614491
ISBN (Print)	9780521844413
DOIs	https://doi.org/10.1017/CBO9780511614491.006
State	Published - Jan 1 2005
Externally published	Yes

All Science Journal Classification (ASJC) codes

General Economics, Econometrics and Finance

Access to Document

10.1017/CBO9780511614491.006

Cite this

@inbook{8a6686f0420b41b9b3b7209fe701ee32,

title = "Testing for weak instruments in Linear Iv regression",

abstract = "Weak instruments can produce biased IV estimators and hypothesis tests with large size distortions. But what, precisely, are weak instruments, and how does one detect them in practice? This paper proposes quantitative definitions of weak instruments based on the maximum IV estimator bias, or the maximumWald test size distortion, when there are multiple endogenous regressors.We tabulate critical values that enable using the first-stage F-statistic (or, when there are multiple endogenous regressors, the Cragg–Donald [1993] statistic) to test whether the given instruments are weak.",

author = "Stock, {James H.} and Motohiro Yogo",

note = "Publisher Copyright: {\textcopyright} Cambridge University Press 2005.",

year = "2005",

month = jan,

day = "1",

doi = "10.1017/CBO9780511614491.006",

language = "English (US)",

isbn = "9780521844413",

pages = "80--108",

booktitle = "Identification and Inference for Econometric Models",

publisher = "Cambridge University Press",

address = "United Kingdom",

}

TY - CHAP

T1 - Testing for weak instruments in Linear Iv regression

AU - Stock, James H.

AU - Yogo, Motohiro

PY - 2005/1/1

Y1 - 2005/1/1

N2 - Weak instruments can produce biased IV estimators and hypothesis tests with large size distortions. But what, precisely, are weak instruments, and how does one detect them in practice? This paper proposes quantitative definitions of weak instruments based on the maximum IV estimator bias, or the maximumWald test size distortion, when there are multiple endogenous regressors.We tabulate critical values that enable using the first-stage F-statistic (or, when there are multiple endogenous regressors, the Cragg–Donald [1993] statistic) to test whether the given instruments are weak.

AB - Weak instruments can produce biased IV estimators and hypothesis tests with large size distortions. But what, precisely, are weak instruments, and how does one detect them in practice? This paper proposes quantitative definitions of weak instruments based on the maximum IV estimator bias, or the maximumWald test size distortion, when there are multiple endogenous regressors.We tabulate critical values that enable using the first-stage F-statistic (or, when there are multiple endogenous regressors, the Cragg–Donald [1993] statistic) to test whether the given instruments are weak.

UR - http://www.scopus.com/inward/record.url?scp=84926100534&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84926100534&partnerID=8YFLogxK

U2 - 10.1017/CBO9780511614491.006

DO - 10.1017/CBO9780511614491.006

M3 - Chapter

AN - SCOPUS:84926100534

SN - 9780521844413

SP - 80

EP - 108

BT - Identification and Inference for Econometric Models

PB - Cambridge University Press

ER -

Testing for weak instruments in Linear Iv regression

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this