I have been reading about Bayesian methods for performing survival analysis. One of the things that struck me is that in the three or so books that I have read, there is no mention of handling ties between events, e.g. Efron, Breslow, or exact.

Is this complication limited to frequentist statistical treatments?

  • 1
    $\begingroup$ What kind of Bayesian survival analysis are you talking about ? Parametric ? Nonparametric ? Semiparametric ? $\endgroup$ – Stéphane Laurent Nov 26 '12 at 22:22
  • $\begingroup$ e.g. Cox Proportional Hazards. I am wondering if it has something to do with the partial likelihood vs. the full likelihood. $\endgroup$ – Eric Brown Nov 26 '12 at 23:18
  • $\begingroup$ Ties also affect Bayesian inference. The presence of ties in a sample may preclude the existence of the posterior distribution even under the use of proper priors (see). $\endgroup$ – user10525 Nov 27 '12 at 9:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.