Using Frailties in the Accelerated Failure Time Model

Research output: Contribution to journalArticlepeer-review

36 Scopus citations


The accelerated failure time (AFT) model is an important alternative to the Cox proportional hazards model (PHM) in survival analysis. For multivariate failure time data we propose to use frailties to explicitly account for possible correlations (and heterogeneity) among failure times. An EM-like algorithm analogous to that in the frailty model for the Cox model is adapted. Through simulation it is shown that its performance compares favorably with that of the marginal independence approach. For illustration we reanalyze a real data set.

Original languageEnglish (US)
Pages (from-to)55-64
Number of pages10
JournalLifetime Data Analysis
Issue number1
StatePublished - Dec 1 2001


  • AFT model
  • Censoring
  • Clustering
  • EM algorithm
  • Frailty model
  • Proportional hazards model


Dive into the research topics of 'Using Frailties in the Accelerated Failure Time Model'. Together they form a unique fingerprint.

Cite this