Testing instrument validity for LATE identification based on inequality moment constraints

Martin Huber, Giovanni Mellace

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Resumé

We derive testable implications of instrument validity in just identified treatment effect models with endogeneity and consider several tests. The identifying assumptions of the local average treatment effect allow us to both point identify and bound the mean potential outcomes (i) of the always takers under treatment and (ii) of the never takers under non-treatment. The point identified means must lie within their respective bounds, which provides us with four testable inequality moment constraints. Finally, we adapt our testing framework to the identification of distributional features. A brief simulation study and an application to labor market data are also provided.
OriginalsprogEngelsk
TidsskriftReview of Economics and Statistics
Vol/bind97
Udgave nummer2
Sider (fra-til)398–411
ISSN0034-6535
DOI
StatusUdgivet - 2015

Fingeraftryk

labor market
simulation
Testing instrument
Moment inequalities
Average treatment effect
Potential outcomes
Market data
Treatment effects
Testing
Labour market
Simulation study
Endogeneity

Citer dette

@article{7830d35acd954bd7a52730e5b4894a06,
title = "Testing instrument validity for LATE identification based on inequality moment constraints",
abstract = "We derive testable implications of instrument validity in just identified treatment effect models with endogeneity and consider several tests. The identifying assumptions of the local average treatment effect allow us to both point identify and bound the mean potential outcomes (i) of the always takers under treatment and (ii) of the never takers under non-treatment. The point identified means must lie within their respective bounds, which provides us with four testable inequality moment constraints. Finally, we adapt our testing framework to the identification of distributional features. A brief simulation study and an application to labor market data are also provided.",
author = "Martin Huber and Giovanni Mellace",
year = "2015",
doi = "10.1162/REST_a_00450",
language = "English",
volume = "97",
pages = "398–411",
journal = "Review of Economics and Statistics",
issn = "0034-6535",
publisher = "Heinemann",
number = "2",

}

Testing instrument validity for LATE identification based on inequality moment constraints. / Huber, Martin; Mellace, Giovanni.

I: Review of Economics and Statistics, Bind 97, Nr. 2, 2015, s. 398–411.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Testing instrument validity for LATE identification based on inequality moment constraints

AU - Huber, Martin

AU - Mellace, Giovanni

PY - 2015

Y1 - 2015

N2 - We derive testable implications of instrument validity in just identified treatment effect models with endogeneity and consider several tests. The identifying assumptions of the local average treatment effect allow us to both point identify and bound the mean potential outcomes (i) of the always takers under treatment and (ii) of the never takers under non-treatment. The point identified means must lie within their respective bounds, which provides us with four testable inequality moment constraints. Finally, we adapt our testing framework to the identification of distributional features. A brief simulation study and an application to labor market data are also provided.

AB - We derive testable implications of instrument validity in just identified treatment effect models with endogeneity and consider several tests. The identifying assumptions of the local average treatment effect allow us to both point identify and bound the mean potential outcomes (i) of the always takers under treatment and (ii) of the never takers under non-treatment. The point identified means must lie within their respective bounds, which provides us with four testable inequality moment constraints. Finally, we adapt our testing framework to the identification of distributional features. A brief simulation study and an application to labor market data are also provided.

U2 - 10.1162/REST_a_00450

DO - 10.1162/REST_a_00450

M3 - Journal article

VL - 97

SP - 398

EP - 411

JO - Review of Economics and Statistics

JF - Review of Economics and Statistics

SN - 0034-6535

IS - 2

ER -