Say I have a lifetime variable with survival probability $S(t|\theta)$, and I estimate this probability by $$\hat{S}(t|\theta)=S(t|\hat{\theta})$$ How would we go about deriving the standard error of this estimate? Is there a general procedure to follow, or does it depend on the form of the hazard function of $T$?

  • 1
    $\begingroup$ It may help to clarify how you would like to estimate $S$: is this the Kaplan Meier estimator you are interested in? Cox-PH model with Kaplan Meier base distribution? Or a fully parametric model? $\endgroup$
    – Cliff AB
    Jan 17, 2017 at 15:20
  • $\begingroup$ I am working with a semiparametric additive model, specifically with hazard $\lambda(t|\alpha,\beta(t))=X^T\beta(t)+g(Y^T\alpha)$ $\endgroup$
    – user128836
    Jan 17, 2017 at 15:39


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.