is it possible to perform a mediation analysis to calculate total, direct, indirect effects obtaining the Hazard Ratio (rather than odds ratio) with Stata?
-
Login or Register
- Log in with
- You are not logged in. You can browse but not post. Login or Register by clicking 'Login or Register' at the top-right of this page. For more information on Statalist, see the FAQ.
-
-
You didn't get a quick answer. You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output and sample data using dataex.
The specific answer would almost certainly depend on exactly what routine you're using to do the mediation analysis. -
I don't know if there are standard tools to do this, but here's an approach: -khb- (available at ssc) does mediation analyses for various regression models, including -clogit- and -cloglog-. If you are doing a survival model in continuous time, you perhaps could exploit the identity of the likelihood for Cox regression to that for the conditional logit and estimate your model with -clogit-. (Google on how to do that.) Or, if you are doing a discrete time model, expanding the data set to person-period and then estimating a -cloglog- model will give hazard ratios. Whether this is a completely defensible way to do a mediation analysis, I don't know. And, if you are thinking of hazard ratios in some other context than a survival model, then of course the preceding is irrelevant.
Comment
Comment