search

kgoldfeld / RDataGenBlog

0 stars 1 forks source link

Question regarding blogpost #2

Open Alik-V opened 2 years ago

Alik-V commented 2 years ago

Hi Keith, Thank you so much for your brilliant simstudy package, it is fantastic tool for data simulation. I am writing with a question about one of the posts in your blog - Follow-up: simstudy function for generating parameters for survival distribution.

When you provided a code example here: image https://github.com/kgoldfeld/RDataGenBlog/blob/b58862b22d2db09085b7e6db0db31a95b165d2aa/content/post/2022-02-22-follow-up-simstudy-function-for-generating-parameters-for-survival-distribution/index.en.Rmd#L137

Is 0.8 coefficient by rx variable essentially equates to a 0.8 hazard ratio?

kgoldfeld commented 2 years ago

Hey Alik -

Thanks for your note. Yes, you are essentially correct - it is interpreted as you would in a Cox proportional hazards model, so it would be the log(HR). At the end of the post, I fit a model, and you can see the parameter is recovered quite closely. The point estimate of the hazard ratio is exp(-0.78) = 0.46, close to the true hazard ratio exp(-0.8) = 0.45.

Let me know if that answers your question, and if you have any other questions.