Model selection in nonparametric hazard regression

Chenlei Leng, Hao Zhang

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

We propose a novel model selection method for a nonparametric extension of the Cox proportional hazard model, in the framework of smoothing splines ANOVA models. The method automates the model building and model selection processes simultaneously by penalizing the reproducing kernel Hilbert space norms. On the basis of a reformulation of the penalized partial likelihood, we propose an efficient algorithm to compute the estimate. The solution demonstrates great flexibility and easy interpretability in modeling relative risk functions for censored data. Adaptive choice of the smoothing parameter is discussed. Both simulations and a real example suggest that our proposal is a useful tool for multivariate function estimation and model selection in survival analysis.

Original languageEnglish (US)
Pages (from-to)417-429
Number of pages13
JournalJournal of Nonparametric Statistics
Volume18
Issue number7-8
DOIs
StatePublished - Oct 2006
Externally publishedYes

Fingerprint

Model Selection
Hazard
Regression
Partial Likelihood
Risk Function
Cox Proportional Hazards Model
Penalized Likelihood
Smoothing Splines
Function Estimation
Reproducing Kernel Hilbert Space
Multivariate Functions
Relative Risk
Survival Analysis
Smoothing Parameter
Interpretability
Censored Data
Reformulation
Efficient Algorithms
Flexibility
Norm

Keywords

  • COSSO
  • Cox proportional hazard model
  • Model selection
  • Penalized likelihood

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

Model selection in nonparametric hazard regression. / Leng, Chenlei; Zhang, Hao.

In: Journal of Nonparametric Statistics, Vol. 18, No. 7-8, 10.2006, p. 417-429.

Research output: Contribution to journalArticle

@article{4caa1e3e5226446dbd6dfdbca5852b8c,
title = "Model selection in nonparametric hazard regression",
abstract = "We propose a novel model selection method for a nonparametric extension of the Cox proportional hazard model, in the framework of smoothing splines ANOVA models. The method automates the model building and model selection processes simultaneously by penalizing the reproducing kernel Hilbert space norms. On the basis of a reformulation of the penalized partial likelihood, we propose an efficient algorithm to compute the estimate. The solution demonstrates great flexibility and easy interpretability in modeling relative risk functions for censored data. Adaptive choice of the smoothing parameter is discussed. Both simulations and a real example suggest that our proposal is a useful tool for multivariate function estimation and model selection in survival analysis.",
keywords = "COSSO, Cox proportional hazard model, Model selection, Penalized likelihood",
author = "Chenlei Leng and Hao Zhang",
year = "2006",
month = "10",
doi = "10.1080/10485250601027042",
language = "English (US)",
volume = "18",
pages = "417--429",
journal = "Journal of Nonparametric Statistics",
issn = "1048-5252",
publisher = "Taylor and Francis Ltd.",
number = "7-8",

}

TY - JOUR

T1 - Model selection in nonparametric hazard regression

AU - Leng, Chenlei

AU - Zhang, Hao

PY - 2006/10

Y1 - 2006/10

N2 - We propose a novel model selection method for a nonparametric extension of the Cox proportional hazard model, in the framework of smoothing splines ANOVA models. The method automates the model building and model selection processes simultaneously by penalizing the reproducing kernel Hilbert space norms. On the basis of a reformulation of the penalized partial likelihood, we propose an efficient algorithm to compute the estimate. The solution demonstrates great flexibility and easy interpretability in modeling relative risk functions for censored data. Adaptive choice of the smoothing parameter is discussed. Both simulations and a real example suggest that our proposal is a useful tool for multivariate function estimation and model selection in survival analysis.

AB - We propose a novel model selection method for a nonparametric extension of the Cox proportional hazard model, in the framework of smoothing splines ANOVA models. The method automates the model building and model selection processes simultaneously by penalizing the reproducing kernel Hilbert space norms. On the basis of a reformulation of the penalized partial likelihood, we propose an efficient algorithm to compute the estimate. The solution demonstrates great flexibility and easy interpretability in modeling relative risk functions for censored data. Adaptive choice of the smoothing parameter is discussed. Both simulations and a real example suggest that our proposal is a useful tool for multivariate function estimation and model selection in survival analysis.

KW - COSSO

KW - Cox proportional hazard model

KW - Model selection

KW - Penalized likelihood

UR - http://www.scopus.com/inward/record.url?scp=34547733991&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547733991&partnerID=8YFLogxK

U2 - 10.1080/10485250601027042

DO - 10.1080/10485250601027042

M3 - Article

AN - SCOPUS:34547733991

VL - 18

SP - 417

EP - 429

JO - Journal of Nonparametric Statistics

JF - Journal of Nonparametric Statistics

SN - 1048-5252

IS - 7-8

ER -