I am trying to simulate the survival data (by using Weibull distribution) that can fit the Cox model below:

h(t) = ho(t). exp(beta1 * X1 + beta2 * X1 * X2)

X1 and X2 are binary.

I haved tried using the linear predictor lp = exp(beta1*X1 + beta2 * X1 * X2) but it seems wrong. When I fitted the Cox model later on, the estimated beta1 and beta2 are absolutely different with the value of beta1 and beta2 that I have used to simulate data.

Someone can help me please? Thankyou so much.

  • 1
    $\begingroup$ Could you post some of the code for your simulation and evidence for the differences you see? $\endgroup$ – Alex R. May 20 '16 at 18:21

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.