I've built a survival model, and I want to examine pairwise comparisons between my treatment groups. I have an advertising data set where I am modeling a conversion event(Purchase = 1, No Purchase = 0), using number of impressions as my time variable and adgroup(with 5 levels - control, seg1, seg2, seg3, and seg,4) as my covariate.

Using a CoxPH model, I've ran some code to get contrasts between seg1 and seg2.

>contrast(imp.cph, list(first_imp_adgroup_id='seg1'),list(first_imp_adgroup_id='seg2'))

   Contrast      S.E.     Lower     Upper    Z Pr(>|z|)
  11 0.574862 0.1344827 0.3112808 0.8384433 4.27        0

Confidence intervals are 0.95 individual intervals

My first question is, how do I interpret that contrast coefficient?

Also, is my model baseline (the control) going to affect my interpretation?

I've never done survival analysis before, but my boss just threw this at me, so I'm trying to learn!


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.