On estimation of partially linear transformation models

Wenbin Lu, Hao Zhang

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

We study a general class of partially linear transformation models, which extend linear transformation models by incorporating nonlinear covariate effects in survival data analysis. A new martingale-based estimating equation approach, consisting of both global and kernelweighted local estimation equations, is developed for estimating the parametric and nonparametric covariate effects in a unified manner. We show that with a proper choice of the kernel bandwidth parameter, one can obtain the consistent and asymptotically normal parameter estimates for the linear effects. Asymptotic properties of the estimated nonlinear effects are established as well.We further suggest a simple resampling method to estimate the asymptotic variance of the linear estimates and show its effectiveness. To facilitate the implementation of the new procedure, an iterative algorithm is developed. Numerical examples are given to illustrate the finite-sample performance of the procedure. Supplementary materials are available online.

Original languageEnglish (US)
Pages (from-to)683-691
Number of pages9
JournalJournal. American Statistical Association
Volume105
Issue number490
DOIs
StatePublished - Jun 2010
Externally publishedYes

Fingerprint

Linear Transformation Model
Covariates
Estimate
Resampling Methods
Estimating Equation
Survival Data
Asymptotic Variance
Nonlinear Effects
Martingale
Iterative Algorithm
Asymptotic Properties
Data analysis
Bandwidth
kernel
Numerical Examples
Transformation model

Keywords

  • Estimating equations
  • Local polynomials
  • Martingale
  • Resampling

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

On estimation of partially linear transformation models. / Lu, Wenbin; Zhang, Hao.

In: Journal. American Statistical Association, Vol. 105, No. 490, 06.2010, p. 683-691.

Research output: Contribution to journalArticle

@article{b8895d9b9855446faaade65eb1a70d3b,
title = "On estimation of partially linear transformation models",
abstract = "We study a general class of partially linear transformation models, which extend linear transformation models by incorporating nonlinear covariate effects in survival data analysis. A new martingale-based estimating equation approach, consisting of both global and kernelweighted local estimation equations, is developed for estimating the parametric and nonparametric covariate effects in a unified manner. We show that with a proper choice of the kernel bandwidth parameter, one can obtain the consistent and asymptotically normal parameter estimates for the linear effects. Asymptotic properties of the estimated nonlinear effects are established as well.We further suggest a simple resampling method to estimate the asymptotic variance of the linear estimates and show its effectiveness. To facilitate the implementation of the new procedure, an iterative algorithm is developed. Numerical examples are given to illustrate the finite-sample performance of the procedure. Supplementary materials are available online.",
keywords = "Estimating equations, Local polynomials, Martingale, Resampling",
author = "Wenbin Lu and Hao Zhang",
year = "2010",
month = "6",
doi = "10.1198/jasa.2010.tm09302",
language = "English (US)",
volume = "105",
pages = "683--691",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "490",

}

TY - JOUR

T1 - On estimation of partially linear transformation models

AU - Lu, Wenbin

AU - Zhang, Hao

PY - 2010/6

Y1 - 2010/6

N2 - We study a general class of partially linear transformation models, which extend linear transformation models by incorporating nonlinear covariate effects in survival data analysis. A new martingale-based estimating equation approach, consisting of both global and kernelweighted local estimation equations, is developed for estimating the parametric and nonparametric covariate effects in a unified manner. We show that with a proper choice of the kernel bandwidth parameter, one can obtain the consistent and asymptotically normal parameter estimates for the linear effects. Asymptotic properties of the estimated nonlinear effects are established as well.We further suggest a simple resampling method to estimate the asymptotic variance of the linear estimates and show its effectiveness. To facilitate the implementation of the new procedure, an iterative algorithm is developed. Numerical examples are given to illustrate the finite-sample performance of the procedure. Supplementary materials are available online.

AB - We study a general class of partially linear transformation models, which extend linear transformation models by incorporating nonlinear covariate effects in survival data analysis. A new martingale-based estimating equation approach, consisting of both global and kernelweighted local estimation equations, is developed for estimating the parametric and nonparametric covariate effects in a unified manner. We show that with a proper choice of the kernel bandwidth parameter, one can obtain the consistent and asymptotically normal parameter estimates for the linear effects. Asymptotic properties of the estimated nonlinear effects are established as well.We further suggest a simple resampling method to estimate the asymptotic variance of the linear estimates and show its effectiveness. To facilitate the implementation of the new procedure, an iterative algorithm is developed. Numerical examples are given to illustrate the finite-sample performance of the procedure. Supplementary materials are available online.

KW - Estimating equations

KW - Local polynomials

KW - Martingale

KW - Resampling

UR - http://www.scopus.com/inward/record.url?scp=78649427504&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78649427504&partnerID=8YFLogxK

U2 - 10.1198/jasa.2010.tm09302

DO - 10.1198/jasa.2010.tm09302

M3 - Article

AN - SCOPUS:78649427504

VL - 105

SP - 683

EP - 691

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 490

ER -