A simple iterative Z-estimator for semiparametric models

Research output: Contribution to journalArticleResearchpeer-review

Abstract

We propose a new iterative estimation algorithm for use in semiparametric models where calculation of Z-estimators by conventional means is difficult or impossible. Unlike a Newton–Raphson approach, which makes use of the entire Hessian, this approach only uses curvature information associated with portions of the Hessian that are relatively easy to calculate. Consistency and asymptotic normality of estimators obtained from this algorithm are established under regularity conditions and an information dominance condition. Two specific examples, a quantile regression model with missing covariates and a GARCH-in-mean model with conditional mean of unknown functional form, demonstrate the applicability of the algorithm. This new approach can be interpreted as an extension of the maximization by parts estimation approach to semiparametric models.

Original languageEnglish
Pages (from-to)111-141
Number of pages31
JournalEconometric Theory
Volume35
Issue number1
DOIs
Publication statusPublished - Feb 2019

Cite this

@article{0772c59b5470421b804853c08b135885,
title = "A simple iterative Z-estimator for semiparametric models",
abstract = "We propose a new iterative estimation algorithm for use in semiparametric models where calculation of Z-estimators by conventional means is difficult or impossible. Unlike a Newton–Raphson approach, which makes use of the entire Hessian, this approach only uses curvature information associated with portions of the Hessian that are relatively easy to calculate. Consistency and asymptotic normality of estimators obtained from this algorithm are established under regularity conditions and an information dominance condition. Two specific examples, a quantile regression model with missing covariates and a GARCH-in-mean model with conditional mean of unknown functional form, demonstrate the applicability of the algorithm. This new approach can be interpreted as an extension of the maximization by parts estimation approach to semiparametric models.",
author = "Frazier, {David T.}",
year = "2019",
month = "2",
doi = "10.1017/S0266466618000063",
language = "English",
volume = "35",
pages = "111--141",
journal = "Econometric Theory",
issn = "0266-4666",
publisher = "Cambridge University Press",
number = "1",

}

A simple iterative Z-estimator for semiparametric models. / Frazier, David T.

In: Econometric Theory, Vol. 35, No. 1, 02.2019, p. 111-141.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A simple iterative Z-estimator for semiparametric models

AU - Frazier, David T.

PY - 2019/2

Y1 - 2019/2

N2 - We propose a new iterative estimation algorithm for use in semiparametric models where calculation of Z-estimators by conventional means is difficult or impossible. Unlike a Newton–Raphson approach, which makes use of the entire Hessian, this approach only uses curvature information associated with portions of the Hessian that are relatively easy to calculate. Consistency and asymptotic normality of estimators obtained from this algorithm are established under regularity conditions and an information dominance condition. Two specific examples, a quantile regression model with missing covariates and a GARCH-in-mean model with conditional mean of unknown functional form, demonstrate the applicability of the algorithm. This new approach can be interpreted as an extension of the maximization by parts estimation approach to semiparametric models.

AB - We propose a new iterative estimation algorithm for use in semiparametric models where calculation of Z-estimators by conventional means is difficult or impossible. Unlike a Newton–Raphson approach, which makes use of the entire Hessian, this approach only uses curvature information associated with portions of the Hessian that are relatively easy to calculate. Consistency and asymptotic normality of estimators obtained from this algorithm are established under regularity conditions and an information dominance condition. Two specific examples, a quantile regression model with missing covariates and a GARCH-in-mean model with conditional mean of unknown functional form, demonstrate the applicability of the algorithm. This new approach can be interpreted as an extension of the maximization by parts estimation approach to semiparametric models.

UR - http://www.scopus.com/inward/record.url?scp=85045262891&partnerID=8YFLogxK

U2 - 10.1017/S0266466618000063

DO - 10.1017/S0266466618000063

M3 - Article

VL - 35

SP - 111

EP - 141

JO - Econometric Theory

JF - Econometric Theory

SN - 0266-4666

IS - 1

ER -