I have 8 variables in my model out of which I have a group which is definitely not a fixed effect. I tried checking the random effect on the basis of the log-likehood test and it seems significant. I am not sure how to determine whether it is a cluster effect or the random effect. Please help me understand these two concepts.

  • $\begingroup$ Can you explain what you mean by a cluster effect and by a random effect in your context? Alternatively, maybe you can add more concrete specifics to your question - it is too vague to answer as it is at the moment. $\endgroup$ – Isabella Ghement Mar 17 '19 at 15:26
  • $\begingroup$ marginal_model <- coxph(surv(time,status)~age +cluster(inst),data = lung) #this is with the cluster effect conditional_model <- coxme(surv(time,status)~ age+ (1|inst),data=lung) #this is the random effect $\endgroup$ – Ramu Mar 17 '19 at 16:27

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.