Bootstrap techniques for proportional hazards models with censored observations

Daniel Zelterman, Chap T. Le, Thomas A. Louis

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


We propose a bootstrap technique for generating pseudo-samples from survival data containing censored observations. This simulation selects a survival time with replacement from the data and then assigns a covariate according to the model of proportional hazards. We also develop a constrained bootstrap technique in which every pseudo-sample has the same distribution of covariate values as does the original, observed data. We use these simulation techniques to estimate the bias and variance of regression coefficients and to approximate the significance levels of goodness-of-fit statistics for testing the assumption of the proportional hazards model.

Original languageEnglish (US)
Pages (from-to)191-199
Number of pages9
JournalStatistics and Computing
Issue number3
StatePublished - Jan 1 1996
Externally publishedYes


  • Goodness-of-fit
  • Simulation
  • Survival analysis

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Statistics and Probability
  • Statistics, Probability and Uncertainty
  • Computational Theory and Mathematics


Dive into the research topics of 'Bootstrap techniques for proportional hazards models with censored observations'. Together they form a unique fingerprint.

Cite this