On the size distortion of tests after an overidentifying restrictions pretest

Patrik Guggenberger, Gitanjali Kumar

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

In the linear instrumental variables model, we provide theoretical and Monte Carlo evidence for the size distortion of a two-stage hypothesis test that uses a test of overidentifying restrictions (OR) in the first stage. We derive a lower bound for the asymptotic size of the two-stage test. The lower bound is given by the asymptotic size of a test that rejects the null hypothesis when two conditions are met: the test of OR used in the first stage does not reject and the test in the second stage rejects. This lower bound can be as large as 1 - ε P, where ε P is the pretest nominal size, for a parameter space that allows for local non-exogeneity of the instruments but rules out weak instruments.

Original languageEnglish (US)
Pages (from-to)1138-1160
Number of pages23
JournalJournal of Applied Econometrics
Volume27
Issue number7
DOIs
StatePublished - Nov 1 2012

Fingerprint

Size distortion
Pre-test
Lower bounds
evidence
Weak instruments
Hypothesis test
Instrumental variables

All Science Journal Classification (ASJC) codes

  • Social Sciences (miscellaneous)
  • Economics and Econometrics

Cite this

@article{efe2f4c2cf1240b0adb04bda66ccc8a6,
title = "On the size distortion of tests after an overidentifying restrictions pretest",
abstract = "In the linear instrumental variables model, we provide theoretical and Monte Carlo evidence for the size distortion of a two-stage hypothesis test that uses a test of overidentifying restrictions (OR) in the first stage. We derive a lower bound for the asymptotic size of the two-stage test. The lower bound is given by the asymptotic size of a test that rejects the null hypothesis when two conditions are met: the test of OR used in the first stage does not reject and the test in the second stage rejects. This lower bound can be as large as 1 - ε P, where ε P is the pretest nominal size, for a parameter space that allows for local non-exogeneity of the instruments but rules out weak instruments.",
author = "Patrik Guggenberger and Gitanjali Kumar",
year = "2012",
month = "11",
day = "1",
doi = "10.1002/jae.1251",
language = "English (US)",
volume = "27",
pages = "1138--1160",
journal = "Journal of Applied Econometrics",
issn = "0883-7252",
publisher = "John Wiley and Sons Ltd",
number = "7",

}

On the size distortion of tests after an overidentifying restrictions pretest. / Guggenberger, Patrik; Kumar, Gitanjali.

In: Journal of Applied Econometrics, Vol. 27, No. 7, 01.11.2012, p. 1138-1160.

Research output: Contribution to journalArticle

TY - JOUR

T1 - On the size distortion of tests after an overidentifying restrictions pretest

AU - Guggenberger, Patrik

AU - Kumar, Gitanjali

PY - 2012/11/1

Y1 - 2012/11/1

N2 - In the linear instrumental variables model, we provide theoretical and Monte Carlo evidence for the size distortion of a two-stage hypothesis test that uses a test of overidentifying restrictions (OR) in the first stage. We derive a lower bound for the asymptotic size of the two-stage test. The lower bound is given by the asymptotic size of a test that rejects the null hypothesis when two conditions are met: the test of OR used in the first stage does not reject and the test in the second stage rejects. This lower bound can be as large as 1 - ε P, where ε P is the pretest nominal size, for a parameter space that allows for local non-exogeneity of the instruments but rules out weak instruments.

AB - In the linear instrumental variables model, we provide theoretical and Monte Carlo evidence for the size distortion of a two-stage hypothesis test that uses a test of overidentifying restrictions (OR) in the first stage. We derive a lower bound for the asymptotic size of the two-stage test. The lower bound is given by the asymptotic size of a test that rejects the null hypothesis when two conditions are met: the test of OR used in the first stage does not reject and the test in the second stage rejects. This lower bound can be as large as 1 - ε P, where ε P is the pretest nominal size, for a parameter space that allows for local non-exogeneity of the instruments but rules out weak instruments.

UR - http://www.scopus.com/inward/record.url?scp=84870368644&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870368644&partnerID=8YFLogxK

U2 - 10.1002/jae.1251

DO - 10.1002/jae.1251

M3 - Article

AN - SCOPUS:84870368644

VL - 27

SP - 1138

EP - 1160

JO - Journal of Applied Econometrics

JF - Journal of Applied Econometrics

SN - 0883-7252

IS - 7

ER -