Pairwise difference estimation with nonparametric control variables

Andres Aradillas-Lopez, Bo E. Honoré, James L. Powell

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

This article extends the pairwise difference estimators for various semilinear limited dependent variable models proposed by Honoré and Powell (Identification and Inference in Econometric Models. Essays in Honor of Thomas Rothenberg Cambridge: Cambridge University Press, 2005) to permit the regressor appearing in the nonparametric component to itself depend upon a conditional expectation that is nonparametrically estimated. This permits the estimation approach to be applied to nonlinear models with sample selectivity and/or endogeneity, in which a "control variable" for selectivity or endogeneity is nonparametrically estimated. We develop the relevant asymptotic theory for the proposed estimators and we illustrate the theory to derive the asymptotic distribution of the estimator for the partially linear logit model.

Original languageEnglish (US)
Pages (from-to)1119-1158
Number of pages40
JournalInternational Economic Review
Volume48
Issue number4
DOIs
StatePublished - Nov 1 2007

Fingerprint

Estimator
Control variable
Selectivity
Endogeneity
Limited dependent variable models
Inference
Asymptotic distribution
Asymptotic theory
Conditional expectation
Logit model
Econometric models

All Science Journal Classification (ASJC) codes

  • Economics and Econometrics

Cite this

Aradillas-Lopez, Andres ; Honoré, Bo E. ; Powell, James L. / Pairwise difference estimation with nonparametric control variables. In: International Economic Review. 2007 ; Vol. 48, No. 4. pp. 1119-1158.
@article{04e7a1fd126c497890513ebf64a017bc,
title = "Pairwise difference estimation with nonparametric control variables",
abstract = "This article extends the pairwise difference estimators for various semilinear limited dependent variable models proposed by Honor{\'e} and Powell (Identification and Inference in Econometric Models. Essays in Honor of Thomas Rothenberg Cambridge: Cambridge University Press, 2005) to permit the regressor appearing in the nonparametric component to itself depend upon a conditional expectation that is nonparametrically estimated. This permits the estimation approach to be applied to nonlinear models with sample selectivity and/or endogeneity, in which a {"}control variable{"} for selectivity or endogeneity is nonparametrically estimated. We develop the relevant asymptotic theory for the proposed estimators and we illustrate the theory to derive the asymptotic distribution of the estimator for the partially linear logit model.",
author = "Andres Aradillas-Lopez and Honor{\'e}, {Bo E.} and Powell, {James L.}",
year = "2007",
month = "11",
day = "1",
doi = "10.1111/j.1468-2354.2007.00457.x",
language = "English (US)",
volume = "48",
pages = "1119--1158",
journal = "International Economic Review",
issn = "0020-6598",
publisher = "Wiley-Blackwell",
number = "4",

}

Pairwise difference estimation with nonparametric control variables. / Aradillas-Lopez, Andres; Honoré, Bo E.; Powell, James L.

In: International Economic Review, Vol. 48, No. 4, 01.11.2007, p. 1119-1158.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Pairwise difference estimation with nonparametric control variables

AU - Aradillas-Lopez, Andres

AU - Honoré, Bo E.

AU - Powell, James L.

PY - 2007/11/1

Y1 - 2007/11/1

N2 - This article extends the pairwise difference estimators for various semilinear limited dependent variable models proposed by Honoré and Powell (Identification and Inference in Econometric Models. Essays in Honor of Thomas Rothenberg Cambridge: Cambridge University Press, 2005) to permit the regressor appearing in the nonparametric component to itself depend upon a conditional expectation that is nonparametrically estimated. This permits the estimation approach to be applied to nonlinear models with sample selectivity and/or endogeneity, in which a "control variable" for selectivity or endogeneity is nonparametrically estimated. We develop the relevant asymptotic theory for the proposed estimators and we illustrate the theory to derive the asymptotic distribution of the estimator for the partially linear logit model.

AB - This article extends the pairwise difference estimators for various semilinear limited dependent variable models proposed by Honoré and Powell (Identification and Inference in Econometric Models. Essays in Honor of Thomas Rothenberg Cambridge: Cambridge University Press, 2005) to permit the regressor appearing in the nonparametric component to itself depend upon a conditional expectation that is nonparametrically estimated. This permits the estimation approach to be applied to nonlinear models with sample selectivity and/or endogeneity, in which a "control variable" for selectivity or endogeneity is nonparametrically estimated. We develop the relevant asymptotic theory for the proposed estimators and we illustrate the theory to derive the asymptotic distribution of the estimator for the partially linear logit model.

UR - http://www.scopus.com/inward/record.url?scp=37149043159&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=37149043159&partnerID=8YFLogxK

U2 - 10.1111/j.1468-2354.2007.00457.x

DO - 10.1111/j.1468-2354.2007.00457.x

M3 - Article

VL - 48

SP - 1119

EP - 1158

JO - International Economic Review

JF - International Economic Review

SN - 0020-6598

IS - 4

ER -