Efficient estimation of semiparametric transformation models for the cumulative incidence of competing risks

Lu Mao, D. Y. Lin

Research output: Research - peer-reviewArticle

  • 1 Citations

Abstract

The cumulative incidence is the probability of failure from the cause of interest over a certain time period in the presence of other risks. A semiparametric regression model proposed by Fine and Gray has become the method of choice for formulating the effects of covariates on the cumulative incidence. Its estimation, however, requires modelling of the censoring distribution and is not statistically efficient. We present a broad class of semiparametric transformation models which extends the Fine and Gray model, and we allow for unknown causes of failure. We derive the non-parametric maximum likelihood estimators and develop simple and fast numerical algorithms using the profile likelihood. We establish the consistency, asymptotic normality and semiparametric efficiency of the non-parametric maximum likelihood estimators. In addition, we construct graphical and numerical procedures to evaluate and select models. Finally, we demonstrate the advantages of the proposed methods over the existing methods through extensive simulation studies and an application to a major study on bone marrow transplantation.

LanguageEnglish (US)
Pages573-587
Number of pages15
JournalJournal of the Royal Statistical Society. Series B: Statistical Methodology
Volume79
Issue number2
DOIs
StatePublished - Mar 1 2017

Fingerprint

Transformation Model
Competing Risks
Efficient Estimation
Semiparametric Model
Incidence
Efficient estimation
Transformation model
Competing risks
Nonparametric Maximum Likelihood Estimator
Maximum likelihood estimator
Semiparametric Efficiency
Semiparametric Regression Model
Grey Model
Profile Likelihood
Transplantation
Censoring
Numerical Procedure
Asymptotic Normality
Bone
Numerical Algorithms

Keywords

  • Censoring
  • Non-parametric maximum likelihood estimation
  • Profile likelihood
  • Proportional hazards
  • Semiparametric efficiency
  • Survival analysis

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

@article{beccbe2b2254442b84dfe329464fd8e9,
title = "Efficient estimation of semiparametric transformation models for the cumulative incidence of competing risks",
abstract = "The cumulative incidence is the probability of failure from the cause of interest over a certain time period in the presence of other risks. A semiparametric regression model proposed by Fine and Gray has become the method of choice for formulating the effects of covariates on the cumulative incidence. Its estimation, however, requires modelling of the censoring distribution and is not statistically efficient. We present a broad class of semiparametric transformation models which extends the Fine and Gray model, and we allow for unknown causes of failure. We derive the non-parametric maximum likelihood estimators and develop simple and fast numerical algorithms using the profile likelihood. We establish the consistency, asymptotic normality and semiparametric efficiency of the non-parametric maximum likelihood estimators. In addition, we construct graphical and numerical procedures to evaluate and select models. Finally, we demonstrate the advantages of the proposed methods over the existing methods through extensive simulation studies and an application to a major study on bone marrow transplantation.",
keywords = "Censoring, Non-parametric maximum likelihood estimation, Profile likelihood, Proportional hazards, Semiparametric efficiency, Survival analysis",
author = "Lu Mao and Lin, {D. Y.}",
year = "2017",
month = "3",
doi = "10.1111/rssb.12177",
volume = "79",
pages = "573--587",
journal = "Journal of the Royal Statistical Society. Series B: Statistical Methodology",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Efficient estimation of semiparametric transformation models for the cumulative incidence of competing risks

AU - Mao,Lu

AU - Lin,D. Y.

PY - 2017/3/1

Y1 - 2017/3/1

N2 - The cumulative incidence is the probability of failure from the cause of interest over a certain time period in the presence of other risks. A semiparametric regression model proposed by Fine and Gray has become the method of choice for formulating the effects of covariates on the cumulative incidence. Its estimation, however, requires modelling of the censoring distribution and is not statistically efficient. We present a broad class of semiparametric transformation models which extends the Fine and Gray model, and we allow for unknown causes of failure. We derive the non-parametric maximum likelihood estimators and develop simple and fast numerical algorithms using the profile likelihood. We establish the consistency, asymptotic normality and semiparametric efficiency of the non-parametric maximum likelihood estimators. In addition, we construct graphical and numerical procedures to evaluate and select models. Finally, we demonstrate the advantages of the proposed methods over the existing methods through extensive simulation studies and an application to a major study on bone marrow transplantation.

AB - The cumulative incidence is the probability of failure from the cause of interest over a certain time period in the presence of other risks. A semiparametric regression model proposed by Fine and Gray has become the method of choice for formulating the effects of covariates on the cumulative incidence. Its estimation, however, requires modelling of the censoring distribution and is not statistically efficient. We present a broad class of semiparametric transformation models which extends the Fine and Gray model, and we allow for unknown causes of failure. We derive the non-parametric maximum likelihood estimators and develop simple and fast numerical algorithms using the profile likelihood. We establish the consistency, asymptotic normality and semiparametric efficiency of the non-parametric maximum likelihood estimators. In addition, we construct graphical and numerical procedures to evaluate and select models. Finally, we demonstrate the advantages of the proposed methods over the existing methods through extensive simulation studies and an application to a major study on bone marrow transplantation.

KW - Censoring

KW - Non-parametric maximum likelihood estimation

KW - Profile likelihood

KW - Proportional hazards

KW - Semiparametric efficiency

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=84963826923&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963826923&partnerID=8YFLogxK

U2 - 10.1111/rssb.12177

DO - 10.1111/rssb.12177

M3 - Article

VL - 79

SP - 573

EP - 587

JO - Journal of the Royal Statistical Society. Series B: Statistical Methodology

T2 - Journal of the Royal Statistical Society. Series B: Statistical Methodology

JF - Journal of the Royal Statistical Society. Series B: Statistical Methodology

SN - 1369-7412

IS - 2

ER -