Assessing gamma frailty models for clustered failure time data

Joanna H. Shih, Thomas Louis

Research output: Contribution to journalArticle

Abstract

Proportional hazards frailty models use a random effect, so called frailty, to construct association for clustered failure time data. It is customary to assume that the random frailty follows a gamma distribution. In this paper, we propose a graphical method for assessing adequacy of the proportional hazards frailty models. In particular, we focus on the assessment of the gamma distribution assumption for the frailties. We calculate the average of the posterior expected frailties at several followup time points and compare it at these time points to 1, the known mean frailty. Large discrepancies indicate lack of fit. To aid in assessing the goodness of fit, we derive and estimate the standard error of the mean of the posterior expected frailties at each time point examined. We give an example to illustrate the proposed methodology and perform sensitivity analysis by simulations.

Original languageEnglish (US)
Pages (from-to)205-220
Number of pages16
JournalLifetime Data Analysis
Volume1
Issue number2
DOIs
StatePublished - Jun 1995
Externally publishedYes

Fingerprint

Failure Time Data
Frailty Model
Frailty
Clustered Data
Hazards
Proportional Hazards Models
Sensitivity analysis
Hazard Models
Proportional Hazards
Gamma distribution
Lack of Fit
Graphical Methods
Standard error
Goodness of fit
Random Effects
Discrepancy
Sensitivity Analysis
Calculate
Methodology
Estimate

Keywords

  • Clustered failure times
  • gamma frailties
  • posterior
  • proportional hazards

ASJC Scopus subject areas

  • Medicine(all)
  • Applied Mathematics

Cite this

Assessing gamma frailty models for clustered failure time data. / Shih, Joanna H.; Louis, Thomas.

In: Lifetime Data Analysis, Vol. 1, No. 2, 06.1995, p. 205-220.

Research output: Contribution to journalArticle

@article{7c4d2ab5f1fa4fdb9c456f605c85d7e5,
title = "Assessing gamma frailty models for clustered failure time data",
abstract = "Proportional hazards frailty models use a random effect, so called frailty, to construct association for clustered failure time data. It is customary to assume that the random frailty follows a gamma distribution. In this paper, we propose a graphical method for assessing adequacy of the proportional hazards frailty models. In particular, we focus on the assessment of the gamma distribution assumption for the frailties. We calculate the average of the posterior expected frailties at several followup time points and compare it at these time points to 1, the known mean frailty. Large discrepancies indicate lack of fit. To aid in assessing the goodness of fit, we derive and estimate the standard error of the mean of the posterior expected frailties at each time point examined. We give an example to illustrate the proposed methodology and perform sensitivity analysis by simulations.",
keywords = "Clustered failure times, gamma frailties, posterior, proportional hazards",
author = "Shih, {Joanna H.} and Thomas Louis",
year = "1995",
month = "6",
doi = "10.1007/BF00985771",
language = "English (US)",
volume = "1",
pages = "205--220",
journal = "Lifetime Data Analysis",
issn = "1380-7870",
publisher = "Springer Netherlands",
number = "2",

}

TY - JOUR

T1 - Assessing gamma frailty models for clustered failure time data

AU - Shih, Joanna H.

AU - Louis, Thomas

PY - 1995/6

Y1 - 1995/6

N2 - Proportional hazards frailty models use a random effect, so called frailty, to construct association for clustered failure time data. It is customary to assume that the random frailty follows a gamma distribution. In this paper, we propose a graphical method for assessing adequacy of the proportional hazards frailty models. In particular, we focus on the assessment of the gamma distribution assumption for the frailties. We calculate the average of the posterior expected frailties at several followup time points and compare it at these time points to 1, the known mean frailty. Large discrepancies indicate lack of fit. To aid in assessing the goodness of fit, we derive and estimate the standard error of the mean of the posterior expected frailties at each time point examined. We give an example to illustrate the proposed methodology and perform sensitivity analysis by simulations.

AB - Proportional hazards frailty models use a random effect, so called frailty, to construct association for clustered failure time data. It is customary to assume that the random frailty follows a gamma distribution. In this paper, we propose a graphical method for assessing adequacy of the proportional hazards frailty models. In particular, we focus on the assessment of the gamma distribution assumption for the frailties. We calculate the average of the posterior expected frailties at several followup time points and compare it at these time points to 1, the known mean frailty. Large discrepancies indicate lack of fit. To aid in assessing the goodness of fit, we derive and estimate the standard error of the mean of the posterior expected frailties at each time point examined. We give an example to illustrate the proposed methodology and perform sensitivity analysis by simulations.

KW - Clustered failure times

KW - gamma frailties

KW - posterior

KW - proportional hazards

UR - http://www.scopus.com/inward/record.url?scp=0029420414&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029420414&partnerID=8YFLogxK

U2 - 10.1007/BF00985771

DO - 10.1007/BF00985771

M3 - Article

C2 - 9385102

AN - SCOPUS:0029420414

VL - 1

SP - 205

EP - 220

JO - Lifetime Data Analysis

JF - Lifetime Data Analysis

SN - 1380-7870

IS - 2

ER -