Variable selection in Measurement error models

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

Measurement error data or errors-in-variable data have been collected in many studies. Natural criterion functions are often unavailable for general functional measurement error models due to the lack of information on the distribution of the unobservable covariates. Typically, the parameter estimation is via solving estimating equations. In addition, the construction of such estimating equations routinely requires solving integral equations, hence the computation is often much more intensive compared with ordinary regression models. Because of these difficulties, traditional best subset variable selection procedures are not applicable, and in the measurement error model context, variable selection remains an unsolved issue. In this paper, we develop a framework for variable selection in measurement error models via penalized estimating equations. We first propose a class of selection procedures for general parametric measurement error models and for general semi-parametric measurement error models, and study the asymptotic properties of the proposed procedures. Then, under certain regularity conditions and with a properly chosen regularization parameter, we demonstrate that the proposed procedure performs as well as an oracle procedure. We assess the finite sample performance via Monte Carlo simulation studies and illustrate the proposed methodology through the empirical analysis of a familiar data set.

Original languageEnglish (US)
Pages (from-to)274-300
Number of pages27
JournalBernoulli
Volume16
Issue number1
DOIs
StatePublished - Feb 1 2010

Fingerprint

Measurement Error Model
Variable Selection
Estimating Equation
Selection Procedures
Errors in Variables
Subset Selection
Empirical Analysis
Regularization Parameter
Regularity Conditions
Measurement Error
Asymptotic Properties
Parameter Estimation
Covariates
Regression Model
Integral Equations
Monte Carlo Simulation
Simulation Study
Methodology
Demonstrate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability

Cite this

@article{d08275617a31442cab12e238209d98b9,
title = "Variable selection in Measurement error models",
abstract = "Measurement error data or errors-in-variable data have been collected in many studies. Natural criterion functions are often unavailable for general functional measurement error models due to the lack of information on the distribution of the unobservable covariates. Typically, the parameter estimation is via solving estimating equations. In addition, the construction of such estimating equations routinely requires solving integral equations, hence the computation is often much more intensive compared with ordinary regression models. Because of these difficulties, traditional best subset variable selection procedures are not applicable, and in the measurement error model context, variable selection remains an unsolved issue. In this paper, we develop a framework for variable selection in measurement error models via penalized estimating equations. We first propose a class of selection procedures for general parametric measurement error models and for general semi-parametric measurement error models, and study the asymptotic properties of the proposed procedures. Then, under certain regularity conditions and with a properly chosen regularization parameter, we demonstrate that the proposed procedure performs as well as an oracle procedure. We assess the finite sample performance via Monte Carlo simulation studies and illustrate the proposed methodology through the empirical analysis of a familiar data set.",
author = "Yanyuan Ma and Runze Li",
year = "2010",
month = "2",
day = "1",
doi = "10.3150/09-BEJ205",
language = "English (US)",
volume = "16",
pages = "274--300",
journal = "Bernoulli",
issn = "1350-7265",
publisher = "International Statistical Institute",
number = "1",

}

Variable selection in Measurement error models. / Ma, Yanyuan; Li, Runze.

In: Bernoulli, Vol. 16, No. 1, 01.02.2010, p. 274-300.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Variable selection in Measurement error models

AU - Ma, Yanyuan

AU - Li, Runze

PY - 2010/2/1

Y1 - 2010/2/1

N2 - Measurement error data or errors-in-variable data have been collected in many studies. Natural criterion functions are often unavailable for general functional measurement error models due to the lack of information on the distribution of the unobservable covariates. Typically, the parameter estimation is via solving estimating equations. In addition, the construction of such estimating equations routinely requires solving integral equations, hence the computation is often much more intensive compared with ordinary regression models. Because of these difficulties, traditional best subset variable selection procedures are not applicable, and in the measurement error model context, variable selection remains an unsolved issue. In this paper, we develop a framework for variable selection in measurement error models via penalized estimating equations. We first propose a class of selection procedures for general parametric measurement error models and for general semi-parametric measurement error models, and study the asymptotic properties of the proposed procedures. Then, under certain regularity conditions and with a properly chosen regularization parameter, we demonstrate that the proposed procedure performs as well as an oracle procedure. We assess the finite sample performance via Monte Carlo simulation studies and illustrate the proposed methodology through the empirical analysis of a familiar data set.

AB - Measurement error data or errors-in-variable data have been collected in many studies. Natural criterion functions are often unavailable for general functional measurement error models due to the lack of information on the distribution of the unobservable covariates. Typically, the parameter estimation is via solving estimating equations. In addition, the construction of such estimating equations routinely requires solving integral equations, hence the computation is often much more intensive compared with ordinary regression models. Because of these difficulties, traditional best subset variable selection procedures are not applicable, and in the measurement error model context, variable selection remains an unsolved issue. In this paper, we develop a framework for variable selection in measurement error models via penalized estimating equations. We first propose a class of selection procedures for general parametric measurement error models and for general semi-parametric measurement error models, and study the asymptotic properties of the proposed procedures. Then, under certain regularity conditions and with a properly chosen regularization parameter, we demonstrate that the proposed procedure performs as well as an oracle procedure. We assess the finite sample performance via Monte Carlo simulation studies and illustrate the proposed methodology through the empirical analysis of a familiar data set.

UR - http://www.scopus.com/inward/record.url?scp=77951125256&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951125256&partnerID=8YFLogxK

U2 - 10.3150/09-BEJ205

DO - 10.3150/09-BEJ205

M3 - Article

AN - SCOPUS:77951125256

VL - 16

SP - 274

EP - 300

JO - Bernoulli

JF - Bernoulli

SN - 1350-7265

IS - 1

ER -