Multi-index regression models with missing covariates at random

Xu Guo, Wangli Xu, Lixing Zhu

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

This paper considers estimation of the semiparametric multi-index model with missing covariates at random. A weighted estimating equation is suggested by invoking the inverse selection probability approach, and estimators of the indices are respectively defined when the selection probability is known in advance, is estimated parametrically and nonparametrically. The consistency is provided. For the single-index model, the large sample properties show that the estimators with both parametric and nonparametric plug-in estimations can play an important role to achieve smaller limiting variances than the estimator with the true selection probability. Simulation studies are carried out to assess the finite sample performance of the proposed estimators. The proposed methods are applied to an AIDS clinical trials dataset to examine which method could be more efficient. A horse colic dataset is also analyzed for illustration.

Original languageEnglish (US)
Pages (from-to)345-363
Number of pages19
JournalJournal of Multivariate Analysis
Volume123
DOIs
StatePublished - Jan 1 2014

Fingerprint

Missing Covariates
Regression Model
Estimator
Weighted Estimating Equations
Single-index Model
Plug-in
Clinical Trials
Limiting
Simulation Study
Regression model
Covariates
Index model

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Numerical Analysis
  • Statistics, Probability and Uncertainty

Cite this

Guo, Xu ; Xu, Wangli ; Zhu, Lixing. / Multi-index regression models with missing covariates at random. In: Journal of Multivariate Analysis. 2014 ; Vol. 123. pp. 345-363.
@article{156eb8ce119a41f08d3abb356d949c63,
title = "Multi-index regression models with missing covariates at random",
abstract = "This paper considers estimation of the semiparametric multi-index model with missing covariates at random. A weighted estimating equation is suggested by invoking the inverse selection probability approach, and estimators of the indices are respectively defined when the selection probability is known in advance, is estimated parametrically and nonparametrically. The consistency is provided. For the single-index model, the large sample properties show that the estimators with both parametric and nonparametric plug-in estimations can play an important role to achieve smaller limiting variances than the estimator with the true selection probability. Simulation studies are carried out to assess the finite sample performance of the proposed estimators. The proposed methods are applied to an AIDS clinical trials dataset to examine which method could be more efficient. A horse colic dataset is also analyzed for illustration.",
author = "Xu Guo and Wangli Xu and Lixing Zhu",
year = "2014",
month = "1",
day = "1",
doi = "10.1016/j.jmva.2013.10.006",
language = "English (US)",
volume = "123",
pages = "345--363",
journal = "Journal of Multivariate Analysis",
issn = "0047-259X",
publisher = "Academic Press Inc.",

}

Multi-index regression models with missing covariates at random. / Guo, Xu; Xu, Wangli; Zhu, Lixing.

In: Journal of Multivariate Analysis, Vol. 123, 01.01.2014, p. 345-363.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Multi-index regression models with missing covariates at random

AU - Guo, Xu

AU - Xu, Wangli

AU - Zhu, Lixing

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper considers estimation of the semiparametric multi-index model with missing covariates at random. A weighted estimating equation is suggested by invoking the inverse selection probability approach, and estimators of the indices are respectively defined when the selection probability is known in advance, is estimated parametrically and nonparametrically. The consistency is provided. For the single-index model, the large sample properties show that the estimators with both parametric and nonparametric plug-in estimations can play an important role to achieve smaller limiting variances than the estimator with the true selection probability. Simulation studies are carried out to assess the finite sample performance of the proposed estimators. The proposed methods are applied to an AIDS clinical trials dataset to examine which method could be more efficient. A horse colic dataset is also analyzed for illustration.

AB - This paper considers estimation of the semiparametric multi-index model with missing covariates at random. A weighted estimating equation is suggested by invoking the inverse selection probability approach, and estimators of the indices are respectively defined when the selection probability is known in advance, is estimated parametrically and nonparametrically. The consistency is provided. For the single-index model, the large sample properties show that the estimators with both parametric and nonparametric plug-in estimations can play an important role to achieve smaller limiting variances than the estimator with the true selection probability. Simulation studies are carried out to assess the finite sample performance of the proposed estimators. The proposed methods are applied to an AIDS clinical trials dataset to examine which method could be more efficient. A horse colic dataset is also analyzed for illustration.

UR - http://www.scopus.com/inward/record.url?scp=84887218465&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887218465&partnerID=8YFLogxK

U2 - 10.1016/j.jmva.2013.10.006

DO - 10.1016/j.jmva.2013.10.006

M3 - Article

AN - SCOPUS:84887218465

VL - 123

SP - 345

EP - 363

JO - Journal of Multivariate Analysis

JF - Journal of Multivariate Analysis

SN - 0047-259X

ER -