Likelihood approaches for proportional likelihood ratio model with right-censored data

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Regression methods for survival data with right censoring have been extensively studied under semiparametric transformation models such as the Cox regression model and the proportional odds model. However, their practical application could be limited because of possible violation of model assumption or lack of ready interpretation for the regression coefficients in some cases. As an alternative, in this paper, the proportional likelihood ratio model introduced by Luo and Tsai is extended to flexibly model the relationship between survival outcome and covariates. This model has a natural connection with many important semiparametric models such as generalized linear model and density ratio model and is closely related to biased sampling problems. Compared with the semiparametric transformation model, the proportional likelihood ratio model is appealing and practical in many ways because of its model flexibility and quite direct clinical interpretation. We present two likelihood approaches for the estimation and inference on the target regression parameters under independent and dependent censoring assumptions. Based on a conditional likelihood approach using uncensored failure times, a numerically simple estimation procedure is developed by maximizing a pairwise pseudo-likelihood. We also develop a full likelihood approach, and the most efficient maximum likelihood estimator is obtained by a profile likelihood. Simulation studies are conducted to assess the finite-sample properties of the proposed estimators and compare the efficiency of the two likelihood approaches. An application to survival data for bone marrow transplantation patients of acute leukemia is provided to illustrate the proposed method and other approaches for handling non-proportionality. The relative merits of these methods are discussed in concluding remarks.

Original languageEnglish (US)
Pages (from-to)2467-2479
Number of pages13
JournalStatistics in Medicine
Volume33
Issue number14
DOIs
StatePublished - Jun 30 2014

Fingerprint

Right-censored Data
Likelihood Ratio
Likelihood
Directly proportional
Semiparametric Model
Survival
Transformation Model
Survival Data
Biased Sampling Problem
Bone Marrow Transplantation
Proportional Hazards Models
Model
Regression
Linear Models
Dependent Censoring
Leukemia
Proportional Odds Model
Cox Regression Model
Conditional Likelihood
Pseudo-likelihood

Keywords

  • conditional likelihood
  • pairwise pseudo-likelihood
  • profile likelihood
  • proportional likelihood ratio model
  • right-censored data

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Likelihood approaches for proportional likelihood ratio model with right-censored data. / Zhu, Hong.

In: Statistics in Medicine, Vol. 33, No. 14, 30.06.2014, p. 2467-2479.

Research output: Contribution to journalArticle

@article{d6711f9a0fb142478af7ee2faa774bdc,
title = "Likelihood approaches for proportional likelihood ratio model with right-censored data",
abstract = "Regression methods for survival data with right censoring have been extensively studied under semiparametric transformation models such as the Cox regression model and the proportional odds model. However, their practical application could be limited because of possible violation of model assumption or lack of ready interpretation for the regression coefficients in some cases. As an alternative, in this paper, the proportional likelihood ratio model introduced by Luo and Tsai is extended to flexibly model the relationship between survival outcome and covariates. This model has a natural connection with many important semiparametric models such as generalized linear model and density ratio model and is closely related to biased sampling problems. Compared with the semiparametric transformation model, the proportional likelihood ratio model is appealing and practical in many ways because of its model flexibility and quite direct clinical interpretation. We present two likelihood approaches for the estimation and inference on the target regression parameters under independent and dependent censoring assumptions. Based on a conditional likelihood approach using uncensored failure times, a numerically simple estimation procedure is developed by maximizing a pairwise pseudo-likelihood. We also develop a full likelihood approach, and the most efficient maximum likelihood estimator is obtained by a profile likelihood. Simulation studies are conducted to assess the finite-sample properties of the proposed estimators and compare the efficiency of the two likelihood approaches. An application to survival data for bone marrow transplantation patients of acute leukemia is provided to illustrate the proposed method and other approaches for handling non-proportionality. The relative merits of these methods are discussed in concluding remarks.",
keywords = "conditional likelihood, pairwise pseudo-likelihood, profile likelihood, proportional likelihood ratio model, right-censored data",
author = "Hong Zhu",
year = "2014",
month = "6",
day = "30",
doi = "10.1002/sim.6105",
language = "English (US)",
volume = "33",
pages = "2467--2479",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "14",

}

TY - JOUR

T1 - Likelihood approaches for proportional likelihood ratio model with right-censored data

AU - Zhu, Hong

PY - 2014/6/30

Y1 - 2014/6/30

N2 - Regression methods for survival data with right censoring have been extensively studied under semiparametric transformation models such as the Cox regression model and the proportional odds model. However, their practical application could be limited because of possible violation of model assumption or lack of ready interpretation for the regression coefficients in some cases. As an alternative, in this paper, the proportional likelihood ratio model introduced by Luo and Tsai is extended to flexibly model the relationship between survival outcome and covariates. This model has a natural connection with many important semiparametric models such as generalized linear model and density ratio model and is closely related to biased sampling problems. Compared with the semiparametric transformation model, the proportional likelihood ratio model is appealing and practical in many ways because of its model flexibility and quite direct clinical interpretation. We present two likelihood approaches for the estimation and inference on the target regression parameters under independent and dependent censoring assumptions. Based on a conditional likelihood approach using uncensored failure times, a numerically simple estimation procedure is developed by maximizing a pairwise pseudo-likelihood. We also develop a full likelihood approach, and the most efficient maximum likelihood estimator is obtained by a profile likelihood. Simulation studies are conducted to assess the finite-sample properties of the proposed estimators and compare the efficiency of the two likelihood approaches. An application to survival data for bone marrow transplantation patients of acute leukemia is provided to illustrate the proposed method and other approaches for handling non-proportionality. The relative merits of these methods are discussed in concluding remarks.

AB - Regression methods for survival data with right censoring have been extensively studied under semiparametric transformation models such as the Cox regression model and the proportional odds model. However, their practical application could be limited because of possible violation of model assumption or lack of ready interpretation for the regression coefficients in some cases. As an alternative, in this paper, the proportional likelihood ratio model introduced by Luo and Tsai is extended to flexibly model the relationship between survival outcome and covariates. This model has a natural connection with many important semiparametric models such as generalized linear model and density ratio model and is closely related to biased sampling problems. Compared with the semiparametric transformation model, the proportional likelihood ratio model is appealing and practical in many ways because of its model flexibility and quite direct clinical interpretation. We present two likelihood approaches for the estimation and inference on the target regression parameters under independent and dependent censoring assumptions. Based on a conditional likelihood approach using uncensored failure times, a numerically simple estimation procedure is developed by maximizing a pairwise pseudo-likelihood. We also develop a full likelihood approach, and the most efficient maximum likelihood estimator is obtained by a profile likelihood. Simulation studies are conducted to assess the finite-sample properties of the proposed estimators and compare the efficiency of the two likelihood approaches. An application to survival data for bone marrow transplantation patients of acute leukemia is provided to illustrate the proposed method and other approaches for handling non-proportionality. The relative merits of these methods are discussed in concluding remarks.

KW - conditional likelihood

KW - pairwise pseudo-likelihood

KW - profile likelihood

KW - proportional likelihood ratio model

KW - right-censored data

UR - http://www.scopus.com/inward/record.url?scp=84922391511&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84922391511&partnerID=8YFLogxK

U2 - 10.1002/sim.6105

DO - 10.1002/sim.6105

M3 - Article

VL - 33

SP - 2467

EP - 2479

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 14

ER -