A Gamma-frailty proportional hazards model for bivariate interval-censored data

Prabhashi W.Withana Gamage, Christopher S. McMahan, Lianming Wang, Wanzhu Tu

Research output: Contribution to journalArticle

Abstract

Correlated survival data naturally arise from many clinical and epidemiological studies. For the analysis of such data, the Gamma-frailty proportional hazards (PH) model is a popular choice because the regression parameters have marginal interpretations and the statistical association between the failure times can be explicitly quantified via Kendall's tau. Despite their popularity, Gamma-frailty PH models for correlated interval-censored data have not received as much attention as analogous models for right-censored data. A Gamma-frailty PH model for bivariate interval-censored data is presented and an easy to implement expectation–maximization (EM) algorithm for model fitting is developed. The proposed model adopts a monotone spline representation for the purposes of approximating the unknown conditional cumulative baseline hazard functions, significantly reducing the number of unknown parameters while retaining modeling flexibility. The EM algorithm was derived from a data augmentation procedure involving latent Poisson random variables. Extensive numerical studies illustrate that the proposed method can provide reliable estimation and valid inference, and is moreover robust to the misspecification of the frailty distribution. To further illustrate its use, the proposed method is used to analyze data from an epidemiological study of sexually transmitted infections.

Original languageEnglish (US)
Pages (from-to)354-366
Number of pages13
JournalComputational Statistics and Data Analysis
Volume128
DOIs
StatePublished - Dec 1 2018

Fingerprint

Interval-censored Data
Frailty
Proportional Hazards Model
Hazards
Correlated Data
Expectation-maximization Algorithm
Kendall's tau
Data Augmentation
Right-censored Data
Hazard Function
Misspecification
Survival Data
Model Fitting
Failure Time
Unknown Parameters
Spline
Infection
Numerical Study
Baseline
Monotone

Keywords

  • EM algorithm
  • Gamma-frailty
  • Interval-censored data
  • Monotone splines
  • Multivariate regression
  • Poisson latent variables
  • Proportional hazards model
  • Survival analysis

ASJC Scopus subject areas

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

A Gamma-frailty proportional hazards model for bivariate interval-censored data. / Gamage, Prabhashi W.Withana; McMahan, Christopher S.; Wang, Lianming; Tu, Wanzhu.

In: Computational Statistics and Data Analysis, Vol. 128, 01.12.2018, p. 354-366.

Research output: Contribution to journalArticle

Gamage, Prabhashi W.Withana ; McMahan, Christopher S. ; Wang, Lianming ; Tu, Wanzhu. / A Gamma-frailty proportional hazards model for bivariate interval-censored data. In: Computational Statistics and Data Analysis. 2018 ; Vol. 128. pp. 354-366.
@article{0b6dbf156a2a4fd0b7feb8c0ee7128bd,
title = "A Gamma-frailty proportional hazards model for bivariate interval-censored data",
abstract = "Correlated survival data naturally arise from many clinical and epidemiological studies. For the analysis of such data, the Gamma-frailty proportional hazards (PH) model is a popular choice because the regression parameters have marginal interpretations and the statistical association between the failure times can be explicitly quantified via Kendall's tau. Despite their popularity, Gamma-frailty PH models for correlated interval-censored data have not received as much attention as analogous models for right-censored data. A Gamma-frailty PH model for bivariate interval-censored data is presented and an easy to implement expectation–maximization (EM) algorithm for model fitting is developed. The proposed model adopts a monotone spline representation for the purposes of approximating the unknown conditional cumulative baseline hazard functions, significantly reducing the number of unknown parameters while retaining modeling flexibility. The EM algorithm was derived from a data augmentation procedure involving latent Poisson random variables. Extensive numerical studies illustrate that the proposed method can provide reliable estimation and valid inference, and is moreover robust to the misspecification of the frailty distribution. To further illustrate its use, the proposed method is used to analyze data from an epidemiological study of sexually transmitted infections.",
keywords = "EM algorithm, Gamma-frailty, Interval-censored data, Monotone splines, Multivariate regression, Poisson latent variables, Proportional hazards model, Survival analysis",
author = "Gamage, {Prabhashi W.Withana} and McMahan, {Christopher S.} and Lianming Wang and Wanzhu Tu",
year = "2018",
month = "12",
day = "1",
doi = "10.1016/j.csda.2018.07.016",
language = "English (US)",
volume = "128",
pages = "354--366",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",

}

TY - JOUR

T1 - A Gamma-frailty proportional hazards model for bivariate interval-censored data

AU - Gamage, Prabhashi W.Withana

AU - McMahan, Christopher S.

AU - Wang, Lianming

AU - Tu, Wanzhu

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Correlated survival data naturally arise from many clinical and epidemiological studies. For the analysis of such data, the Gamma-frailty proportional hazards (PH) model is a popular choice because the regression parameters have marginal interpretations and the statistical association between the failure times can be explicitly quantified via Kendall's tau. Despite their popularity, Gamma-frailty PH models for correlated interval-censored data have not received as much attention as analogous models for right-censored data. A Gamma-frailty PH model for bivariate interval-censored data is presented and an easy to implement expectation–maximization (EM) algorithm for model fitting is developed. The proposed model adopts a monotone spline representation for the purposes of approximating the unknown conditional cumulative baseline hazard functions, significantly reducing the number of unknown parameters while retaining modeling flexibility. The EM algorithm was derived from a data augmentation procedure involving latent Poisson random variables. Extensive numerical studies illustrate that the proposed method can provide reliable estimation and valid inference, and is moreover robust to the misspecification of the frailty distribution. To further illustrate its use, the proposed method is used to analyze data from an epidemiological study of sexually transmitted infections.

AB - Correlated survival data naturally arise from many clinical and epidemiological studies. For the analysis of such data, the Gamma-frailty proportional hazards (PH) model is a popular choice because the regression parameters have marginal interpretations and the statistical association between the failure times can be explicitly quantified via Kendall's tau. Despite their popularity, Gamma-frailty PH models for correlated interval-censored data have not received as much attention as analogous models for right-censored data. A Gamma-frailty PH model for bivariate interval-censored data is presented and an easy to implement expectation–maximization (EM) algorithm for model fitting is developed. The proposed model adopts a monotone spline representation for the purposes of approximating the unknown conditional cumulative baseline hazard functions, significantly reducing the number of unknown parameters while retaining modeling flexibility. The EM algorithm was derived from a data augmentation procedure involving latent Poisson random variables. Extensive numerical studies illustrate that the proposed method can provide reliable estimation and valid inference, and is moreover robust to the misspecification of the frailty distribution. To further illustrate its use, the proposed method is used to analyze data from an epidemiological study of sexually transmitted infections.

KW - EM algorithm

KW - Gamma-frailty

KW - Interval-censored data

KW - Monotone splines

KW - Multivariate regression

KW - Poisson latent variables

KW - Proportional hazards model

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=85051674996&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051674996&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2018.07.016

DO - 10.1016/j.csda.2018.07.016

M3 - Article

VL - 128

SP - 354

EP - 366

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

ER -