Characterization of the asymptotic distribution of semiparametric M-estimators

Hidehiko Ichimura, Sokbae Lee

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

This paper develops a concrete formula for the asymptotic distribution of two-step, possibly non-smooth semiparametric M-estimators under general misspecification. Our regularity conditions are relatively straightforward to verify and also weaker than those available in the literature. The first-stage nonparametric estimation may depend on finite dimensional parameters. We characterize: (1) conditions under which the first-stage estimation of nonparametric components do not affect the asymptotic distribution, (2) conditions under which the asymptotic distribution is affected by the derivatives of the first-stage nonparametric estimator with respect to the finite-dimensional parameters, and (3) conditions under which one can allow non-smooth objective functions. Our framework is illustrated by applying it to three examples: (1) profiled estimation of a single index quantile regression model, (2) semiparametric least squares estimation under model misspecification, and (3) a smoothed matching estimator.

Original languageEnglish (US)
Pages (from-to)252-266
Number of pages15
JournalJournal of Econometrics
Volume159
Issue number2
DOIs
StatePublished - Dec 2010
Externally publishedYes

Fingerprint

Asymptotic distribution
M-estimator
Regression model
Regularity
Quantile regression
Model misspecification
Least squares
Nonparametric estimation
Estimator
Misspecification
Derivatives
Matching estimators
Objective function

Keywords

  • Semiparametric estimation
  • Two-step estimators

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Characterization of the asymptotic distribution of semiparametric M-estimators. / Ichimura, Hidehiko; Lee, Sokbae.

In: Journal of Econometrics, Vol. 159, No. 2, 12.2010, p. 252-266.

Research output: Contribution to journalArticle

@article{523bee06441446db86eb277f332b44a2,
title = "Characterization of the asymptotic distribution of semiparametric M-estimators",
abstract = "This paper develops a concrete formula for the asymptotic distribution of two-step, possibly non-smooth semiparametric M-estimators under general misspecification. Our regularity conditions are relatively straightforward to verify and also weaker than those available in the literature. The first-stage nonparametric estimation may depend on finite dimensional parameters. We characterize: (1) conditions under which the first-stage estimation of nonparametric components do not affect the asymptotic distribution, (2) conditions under which the asymptotic distribution is affected by the derivatives of the first-stage nonparametric estimator with respect to the finite-dimensional parameters, and (3) conditions under which one can allow non-smooth objective functions. Our framework is illustrated by applying it to three examples: (1) profiled estimation of a single index quantile regression model, (2) semiparametric least squares estimation under model misspecification, and (3) a smoothed matching estimator.",
keywords = "Semiparametric estimation, Two-step estimators",
author = "Hidehiko Ichimura and Sokbae Lee",
year = "2010",
month = "12",
doi = "10.1016/j.jeconom.2010.05.005",
language = "English (US)",
volume = "159",
pages = "252--266",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "2",

}

TY - JOUR

T1 - Characterization of the asymptotic distribution of semiparametric M-estimators

AU - Ichimura, Hidehiko

AU - Lee, Sokbae

PY - 2010/12

Y1 - 2010/12

N2 - This paper develops a concrete formula for the asymptotic distribution of two-step, possibly non-smooth semiparametric M-estimators under general misspecification. Our regularity conditions are relatively straightforward to verify and also weaker than those available in the literature. The first-stage nonparametric estimation may depend on finite dimensional parameters. We characterize: (1) conditions under which the first-stage estimation of nonparametric components do not affect the asymptotic distribution, (2) conditions under which the asymptotic distribution is affected by the derivatives of the first-stage nonparametric estimator with respect to the finite-dimensional parameters, and (3) conditions under which one can allow non-smooth objective functions. Our framework is illustrated by applying it to three examples: (1) profiled estimation of a single index quantile regression model, (2) semiparametric least squares estimation under model misspecification, and (3) a smoothed matching estimator.

AB - This paper develops a concrete formula for the asymptotic distribution of two-step, possibly non-smooth semiparametric M-estimators under general misspecification. Our regularity conditions are relatively straightforward to verify and also weaker than those available in the literature. The first-stage nonparametric estimation may depend on finite dimensional parameters. We characterize: (1) conditions under which the first-stage estimation of nonparametric components do not affect the asymptotic distribution, (2) conditions under which the asymptotic distribution is affected by the derivatives of the first-stage nonparametric estimator with respect to the finite-dimensional parameters, and (3) conditions under which one can allow non-smooth objective functions. Our framework is illustrated by applying it to three examples: (1) profiled estimation of a single index quantile regression model, (2) semiparametric least squares estimation under model misspecification, and (3) a smoothed matching estimator.

KW - Semiparametric estimation

KW - Two-step estimators

UR - http://www.scopus.com/inward/record.url?scp=77958019518&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77958019518&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2010.05.005

DO - 10.1016/j.jeconom.2010.05.005

M3 - Article

AN - SCOPUS:77958019518

VL - 159

SP - 252

EP - 266

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 2

ER -