Testing overidentifying restrictions with many instruments and heteroskedasticity

John C. Chao, Jerry A. Hausman, Whitney K. Newey, Norman R. Swanson, Tiemen M Woutersen

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

This paper gives a test of overidentifying restrictions that is robust to many instruments and heteroskedasticity. It is based on a jackknife version of the overidentifying test statistic. Correct asymptotic critical values are derived for this statistic when the number of instruments grows large, at a rate up to the sample size. It is also shown that the test is valid when the number of instruments is fixed and there is homoskedasticity. This test improves on recently proposed tests by allowing for heteroskedasticity and by avoiding assumptions on the instrument projection matrix. This paper finds in Monte Carlo studies that the test is more accurate and less sensitive to the number of instruments than the Hausman-Sargan or GMM tests of overidentifying restrictions.

Original languageEnglish (US)
Pages (from-to)15-21
Number of pages7
JournalJournal of Econometrics
Volume178
Issue numberPART 1
DOIs
StatePublished - Jan 2014
Externally publishedYes

Fingerprint

Heteroskedasticity
Restriction
Testing
Statistics
Projection Matrix
Jackknife
Monte Carlo Study
Test Statistic
Statistic
Critical value
Sample Size
Valid

Keywords

  • Heteroskedasticity
  • Instrumental variables
  • Many instruments
  • Overidentification tests
  • Specifications tests
  • Weak instruments

ASJC Scopus subject areas

  • Economics and Econometrics
  • Applied Mathematics
  • History and Philosophy of Science

Cite this

Testing overidentifying restrictions with many instruments and heteroskedasticity. / Chao, John C.; Hausman, Jerry A.; Newey, Whitney K.; Swanson, Norman R.; Woutersen, Tiemen M.

In: Journal of Econometrics, Vol. 178, No. PART 1, 01.2014, p. 15-21.

Research output: Contribution to journalArticle

Chao, John C. ; Hausman, Jerry A. ; Newey, Whitney K. ; Swanson, Norman R. ; Woutersen, Tiemen M. / Testing overidentifying restrictions with many instruments and heteroskedasticity. In: Journal of Econometrics. 2014 ; Vol. 178, No. PART 1. pp. 15-21.
@article{5b5c9212ea8741a1a796b3c0fd03d162,
title = "Testing overidentifying restrictions with many instruments and heteroskedasticity",
abstract = "This paper gives a test of overidentifying restrictions that is robust to many instruments and heteroskedasticity. It is based on a jackknife version of the overidentifying test statistic. Correct asymptotic critical values are derived for this statistic when the number of instruments grows large, at a rate up to the sample size. It is also shown that the test is valid when the number of instruments is fixed and there is homoskedasticity. This test improves on recently proposed tests by allowing for heteroskedasticity and by avoiding assumptions on the instrument projection matrix. This paper finds in Monte Carlo studies that the test is more accurate and less sensitive to the number of instruments than the Hausman-Sargan or GMM tests of overidentifying restrictions.",
keywords = "Heteroskedasticity, Instrumental variables, Many instruments, Overidentification tests, Specifications tests, Weak instruments",
author = "Chao, {John C.} and Hausman, {Jerry A.} and Newey, {Whitney K.} and Swanson, {Norman R.} and Woutersen, {Tiemen M}",
year = "2014",
month = "1",
doi = "10.1016/j.jeconom.2013.08.003",
language = "English (US)",
volume = "178",
pages = "15--21",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "PART 1",

}

TY - JOUR

T1 - Testing overidentifying restrictions with many instruments and heteroskedasticity

AU - Chao, John C.

AU - Hausman, Jerry A.

AU - Newey, Whitney K.

AU - Swanson, Norman R.

AU - Woutersen, Tiemen M

PY - 2014/1

Y1 - 2014/1

N2 - This paper gives a test of overidentifying restrictions that is robust to many instruments and heteroskedasticity. It is based on a jackknife version of the overidentifying test statistic. Correct asymptotic critical values are derived for this statistic when the number of instruments grows large, at a rate up to the sample size. It is also shown that the test is valid when the number of instruments is fixed and there is homoskedasticity. This test improves on recently proposed tests by allowing for heteroskedasticity and by avoiding assumptions on the instrument projection matrix. This paper finds in Monte Carlo studies that the test is more accurate and less sensitive to the number of instruments than the Hausman-Sargan or GMM tests of overidentifying restrictions.

AB - This paper gives a test of overidentifying restrictions that is robust to many instruments and heteroskedasticity. It is based on a jackknife version of the overidentifying test statistic. Correct asymptotic critical values are derived for this statistic when the number of instruments grows large, at a rate up to the sample size. It is also shown that the test is valid when the number of instruments is fixed and there is homoskedasticity. This test improves on recently proposed tests by allowing for heteroskedasticity and by avoiding assumptions on the instrument projection matrix. This paper finds in Monte Carlo studies that the test is more accurate and less sensitive to the number of instruments than the Hausman-Sargan or GMM tests of overidentifying restrictions.

KW - Heteroskedasticity

KW - Instrumental variables

KW - Many instruments

KW - Overidentification tests

KW - Specifications tests

KW - Weak instruments

UR - http://www.scopus.com/inward/record.url?scp=84889639229&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84889639229&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2013.08.003

DO - 10.1016/j.jeconom.2013.08.003

M3 - Article

AN - SCOPUS:84889639229

VL - 178

SP - 15

EP - 21

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - PART 1

ER -