Identifiability of cure models

Chin-Shang Li, Jeremy M G Taylor, Judy P. Sy

Research output: Contribution to journalArticle

93 Citations (Scopus)

Abstract

Cure models can be used for censored survival data in which a fraction of the observations do not exhibit the event of interest despite long-term follow-up. In this paper we investigate the identifiability of two forms of the cure model, a standard cure model based on a mixture distribution and a non-mixture proportional hazards (PH) model with long-term survivors. In the standard cure model, except for the case where the conditional survival function is independent of covariates and the mixture probability is an arbitrary function of a covariate we show that the parameters of the standard cure model are identified. In the non-mixture PH model, we show the model is identifiable if the distribution function is specified.

Original languageEnglish (US)
Pages (from-to)389-395
Number of pages7
JournalStatistics and Probability Letters
Volume54
Issue number4
DOIs
StatePublished - Oct 15 2001
Externally publishedYes

Fingerprint

Cure Model
Identifiability
Standard Model
Proportional Hazards Model
Covariates
Censored Survival Data
Mixture Distribution
Survival Function
Distribution Function
Model-based
Arbitrary

Keywords

  • Cure model
  • Latency
  • Logistic-Kaplan-Meier model
  • Logistic-proportional hazards model
  • Long-term incidence

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this

Identifiability of cure models. / Li, Chin-Shang; Taylor, Jeremy M G; Sy, Judy P.

In: Statistics and Probability Letters, Vol. 54, No. 4, 15.10.2001, p. 389-395.

Research output: Contribution to journalArticle

Li, Chin-Shang ; Taylor, Jeremy M G ; Sy, Judy P. / Identifiability of cure models. In: Statistics and Probability Letters. 2001 ; Vol. 54, No. 4. pp. 389-395.
@article{87ca0e87605444cbb08bcc3f5ddf2212,
title = "Identifiability of cure models",
abstract = "Cure models can be used for censored survival data in which a fraction of the observations do not exhibit the event of interest despite long-term follow-up. In this paper we investigate the identifiability of two forms of the cure model, a standard cure model based on a mixture distribution and a non-mixture proportional hazards (PH) model with long-term survivors. In the standard cure model, except for the case where the conditional survival function is independent of covariates and the mixture probability is an arbitrary function of a covariate we show that the parameters of the standard cure model are identified. In the non-mixture PH model, we show the model is identifiable if the distribution function is specified.",
keywords = "Cure model, Latency, Logistic-Kaplan-Meier model, Logistic-proportional hazards model, Long-term incidence",
author = "Chin-Shang Li and Taylor, {Jeremy M G} and Sy, {Judy P.}",
year = "2001",
month = "10",
day = "15",
doi = "10.1016/S0167-7152(01)00105-5",
language = "English (US)",
volume = "54",
pages = "389--395",
journal = "Statistics and Probability Letters",
issn = "0167-7152",
publisher = "Elsevier",
number = "4",

}

TY - JOUR

T1 - Identifiability of cure models

AU - Li, Chin-Shang

AU - Taylor, Jeremy M G

AU - Sy, Judy P.

PY - 2001/10/15

Y1 - 2001/10/15

N2 - Cure models can be used for censored survival data in which a fraction of the observations do not exhibit the event of interest despite long-term follow-up. In this paper we investigate the identifiability of two forms of the cure model, a standard cure model based on a mixture distribution and a non-mixture proportional hazards (PH) model with long-term survivors. In the standard cure model, except for the case where the conditional survival function is independent of covariates and the mixture probability is an arbitrary function of a covariate we show that the parameters of the standard cure model are identified. In the non-mixture PH model, we show the model is identifiable if the distribution function is specified.

AB - Cure models can be used for censored survival data in which a fraction of the observations do not exhibit the event of interest despite long-term follow-up. In this paper we investigate the identifiability of two forms of the cure model, a standard cure model based on a mixture distribution and a non-mixture proportional hazards (PH) model with long-term survivors. In the standard cure model, except for the case where the conditional survival function is independent of covariates and the mixture probability is an arbitrary function of a covariate we show that the parameters of the standard cure model are identified. In the non-mixture PH model, we show the model is identifiable if the distribution function is specified.

KW - Cure model

KW - Latency

KW - Logistic-Kaplan-Meier model

KW - Logistic-proportional hazards model

KW - Long-term incidence

UR - http://www.scopus.com/inward/record.url?scp=0005091283&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0005091283&partnerID=8YFLogxK

U2 - 10.1016/S0167-7152(01)00105-5

DO - 10.1016/S0167-7152(01)00105-5

M3 - Article

VL - 54

SP - 389

EP - 395

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

SN - 0167-7152

IS - 4

ER -