Is it sensible and possible to cross-validate a survival model? Does it depend on whether there is censoring? If not, why? If the answer depends on the model, then answer for common survival models like a weibull or log-logistic model.

  • $\begingroup$ I've asked a similar question at stats.stackexchange.com/questions/168809/… - my current approach is a zero-one loss on each observation. You could do the same thing on your survival model: take the censoring or death time for each subject and ask your survival model to predict whether the subject is alive or dead at that point, and score it zero-one. $\endgroup$ – gwern Sep 5 '15 at 21:19

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.