Dimension reduction with missing response at random

Xu Guo, Tao Wang, Wangli Xu, Lixing Zhu

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

When there are many predictors, how to efficiently impute responses missing at random is an important problem to deal with for regression analysis because this missing mechanism, unlike missing completely at random, is highly related to high-dimensional predictor vectors. In sufficient dimension reduction framework, the fusion-refinement (FR) method in the literature is a promising approach. To make estimation more accurate and efficient, two methods are suggested in this paper. Among them, one method uses the observed data to help on missing data generation, and the other one is an ad hoc approach that mainly reduces the dimension in the nonparametric smoothing in data generation. A data-adaptive synthesization of these two methods is also developed. Simulations are conducted to examine their performance and a HIV clinical trial dataset is analyzed for illustration.

Original languageEnglish (US)
Pages (from-to)228-242
Number of pages15
JournalComputational Statistics and Data Analysis
Volume69
DOIs
StatePublished - Jan 1 2014

Fingerprint

Dimension Reduction
Regression analysis
Fusion reactions
Predictors
Sufficient Dimension Reduction
Nonparametric Smoothing
Missing Completely at Random
Missing at Random
Missing Data
Regression Analysis
Clinical Trials
Fusion
Refinement
High-dimensional
Simulation

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

Guo, Xu ; Wang, Tao ; Xu, Wangli ; Zhu, Lixing. / Dimension reduction with missing response at random. In: Computational Statistics and Data Analysis. 2014 ; Vol. 69. pp. 228-242.
@article{5e7d830a0e084ed291451aec36cf3992,
title = "Dimension reduction with missing response at random",
abstract = "When there are many predictors, how to efficiently impute responses missing at random is an important problem to deal with for regression analysis because this missing mechanism, unlike missing completely at random, is highly related to high-dimensional predictor vectors. In sufficient dimension reduction framework, the fusion-refinement (FR) method in the literature is a promising approach. To make estimation more accurate and efficient, two methods are suggested in this paper. Among them, one method uses the observed data to help on missing data generation, and the other one is an ad hoc approach that mainly reduces the dimension in the nonparametric smoothing in data generation. A data-adaptive synthesization of these two methods is also developed. Simulations are conducted to examine their performance and a HIV clinical trial dataset is analyzed for illustration.",
author = "Xu Guo and Tao Wang and Wangli Xu and Lixing Zhu",
year = "2014",
month = "1",
day = "1",
doi = "10.1016/j.csda.2013.08.001",
language = "English (US)",
volume = "69",
pages = "228--242",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",

}

Dimension reduction with missing response at random. / Guo, Xu; Wang, Tao; Xu, Wangli; Zhu, Lixing.

In: Computational Statistics and Data Analysis, Vol. 69, 01.01.2014, p. 228-242.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dimension reduction with missing response at random

AU - Guo, Xu

AU - Wang, Tao

AU - Xu, Wangli

AU - Zhu, Lixing

PY - 2014/1/1

Y1 - 2014/1/1

N2 - When there are many predictors, how to efficiently impute responses missing at random is an important problem to deal with for regression analysis because this missing mechanism, unlike missing completely at random, is highly related to high-dimensional predictor vectors. In sufficient dimension reduction framework, the fusion-refinement (FR) method in the literature is a promising approach. To make estimation more accurate and efficient, two methods are suggested in this paper. Among them, one method uses the observed data to help on missing data generation, and the other one is an ad hoc approach that mainly reduces the dimension in the nonparametric smoothing in data generation. A data-adaptive synthesization of these two methods is also developed. Simulations are conducted to examine their performance and a HIV clinical trial dataset is analyzed for illustration.

AB - When there are many predictors, how to efficiently impute responses missing at random is an important problem to deal with for regression analysis because this missing mechanism, unlike missing completely at random, is highly related to high-dimensional predictor vectors. In sufficient dimension reduction framework, the fusion-refinement (FR) method in the literature is a promising approach. To make estimation more accurate and efficient, two methods are suggested in this paper. Among them, one method uses the observed data to help on missing data generation, and the other one is an ad hoc approach that mainly reduces the dimension in the nonparametric smoothing in data generation. A data-adaptive synthesization of these two methods is also developed. Simulations are conducted to examine their performance and a HIV clinical trial dataset is analyzed for illustration.

UR - http://www.scopus.com/inward/record.url?scp=84884274207&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884274207&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2013.08.001

DO - 10.1016/j.csda.2013.08.001

M3 - Article

AN - SCOPUS:84884274207

VL - 69

SP - 228

EP - 242

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

ER -