A nonparametric survival function estimator via censored kernel quantile regressions

Seung Jun Shin, Hao Helen Zhang, Yichao Wu

Research output: Research - peer-reviewArticle

Abstract

In survival data analysis, a central interest is to identify the relationship between a possibly censored survival time and explanatory covariates. In this article, a new censored quantile regression method is proposed and studied in the framework of reproducing kernel Hilbert spaces (RKHS). We first establish the joint piecewise linearity of the regression parameters as a function of regularization parameter λ and quantile level τ. An efficient algorithm is then developed to compute the entire two-dimensional solution surface over the (λ × τ )-plane. Finally, a piecewise linear conditional survival function estimator is constructed based on the solution surface. The method provides a new and flexible survival function estimator without requiring such rigid model assumptions as linearity of the survival time or proportionality of the hazards. One important advantage of the estimator is that it can handle moderately high-dimensional covariates. We carry out an asymptotic analysis to justify the proposed method theoretically, and numerical results are shown to illustrate its competitive finite-sample performance under various simulated scenarios and real applications.

LanguageEnglish (US)
Pages457-478
Number of pages22
JournalStatistica Sinica
Volume27
Issue number1
DOIs
StatePublished - Jan 1 2017

Fingerprint

Kernel Regression
Quantile Regression
Survival Function
Estimator
Quantile regression
Kernel
Survival Time
Linearity
Covariates
Censored Regression
Reproducing Kernel Hilbert Space
Survival Data
Regularization Parameter
Quantile
Asymptotic Analysis
Piecewise Linear
Hazard
Justify
Data analysis
High-dimensional

Keywords

  • Censored kernel quantile regression
  • Conditional survival function
  • Solution surface

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

A nonparametric survival function estimator via censored kernel quantile regressions. / Shin, Seung Jun; Zhang, Hao Helen; Wu, Yichao.

In: Statistica Sinica, Vol. 27, No. 1, 01.01.2017, p. 457-478.

Research output: Research - peer-reviewArticle

@article{538791d31d4b4280b4c20f9db8fe21ed,
title = "A nonparametric survival function estimator via censored kernel quantile regressions",
abstract = "In survival data analysis, a central interest is to identify the relationship between a possibly censored survival time and explanatory covariates. In this article, a new censored quantile regression method is proposed and studied in the framework of reproducing kernel Hilbert spaces (RKHS). We first establish the joint piecewise linearity of the regression parameters as a function of regularization parameter λ and quantile level τ. An efficient algorithm is then developed to compute the entire two-dimensional solution surface over the (λ × τ )-plane. Finally, a piecewise linear conditional survival function estimator is constructed based on the solution surface. The method provides a new and flexible survival function estimator without requiring such rigid model assumptions as linearity of the survival time or proportionality of the hazards. One important advantage of the estimator is that it can handle moderately high-dimensional covariates. We carry out an asymptotic analysis to justify the proposed method theoretically, and numerical results are shown to illustrate its competitive finite-sample performance under various simulated scenarios and real applications.",
keywords = "Censored kernel quantile regression, Conditional survival function, Solution surface",
author = "Shin, {Seung Jun} and Zhang, {Hao Helen} and Yichao Wu",
year = "2017",
month = "1",
doi = "10.5705/ss.202014.0071",
volume = "27",
pages = "457--478",
journal = "Statistica Sinica",
issn = "1017-0405",
publisher = "Institute of Statistical Science",
number = "1",

}

TY - JOUR

T1 - A nonparametric survival function estimator via censored kernel quantile regressions

AU - Shin,Seung Jun

AU - Zhang,Hao Helen

AU - Wu,Yichao

PY - 2017/1/1

Y1 - 2017/1/1

N2 - In survival data analysis, a central interest is to identify the relationship between a possibly censored survival time and explanatory covariates. In this article, a new censored quantile regression method is proposed and studied in the framework of reproducing kernel Hilbert spaces (RKHS). We first establish the joint piecewise linearity of the regression parameters as a function of regularization parameter λ and quantile level τ. An efficient algorithm is then developed to compute the entire two-dimensional solution surface over the (λ × τ )-plane. Finally, a piecewise linear conditional survival function estimator is constructed based on the solution surface. The method provides a new and flexible survival function estimator without requiring such rigid model assumptions as linearity of the survival time or proportionality of the hazards. One important advantage of the estimator is that it can handle moderately high-dimensional covariates. We carry out an asymptotic analysis to justify the proposed method theoretically, and numerical results are shown to illustrate its competitive finite-sample performance under various simulated scenarios and real applications.

AB - In survival data analysis, a central interest is to identify the relationship between a possibly censored survival time and explanatory covariates. In this article, a new censored quantile regression method is proposed and studied in the framework of reproducing kernel Hilbert spaces (RKHS). We first establish the joint piecewise linearity of the regression parameters as a function of regularization parameter λ and quantile level τ. An efficient algorithm is then developed to compute the entire two-dimensional solution surface over the (λ × τ )-plane. Finally, a piecewise linear conditional survival function estimator is constructed based on the solution surface. The method provides a new and flexible survival function estimator without requiring such rigid model assumptions as linearity of the survival time or proportionality of the hazards. One important advantage of the estimator is that it can handle moderately high-dimensional covariates. We carry out an asymptotic analysis to justify the proposed method theoretically, and numerical results are shown to illustrate its competitive finite-sample performance under various simulated scenarios and real applications.

KW - Censored kernel quantile regression

KW - Conditional survival function

KW - Solution surface

UR - http://www.scopus.com/inward/record.url?scp=85011343860&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011343860&partnerID=8YFLogxK

U2 - 10.5705/ss.202014.0071

DO - 10.5705/ss.202014.0071

M3 - Article

VL - 27

SP - 457

EP - 478

JO - Statistica Sinica

T2 - Statistica Sinica

JF - Statistica Sinica

SN - 1017-0405

IS - 1

ER -