[R] Linear relative rate / excess relative risk models
Therneau, Terry M., Ph.D.
therneau at mayo.edu
Thu Aug 28 23:03:31 CEST 2014
On 07/30/2014 05:00 AM, r-help-request at r-project.org wrote:
> A while ago, I inquired about fitting excess relative risk models in R. This is a follow-up about what I ended up doing in case the question pops up again.
> While I was not successful in using standard tools, switching to Bayesian modeling using rstan (mc-stan.org/rstan.html) worked better. The results closely match those from Epicure.
> Using the data here:http://dwoll.de/err/dat.txt
> The stan model fit below replicates the results from Epicure here:http://dwoll.de/err/epicure.log
> Of course I am still interested in learning about other options or approaches.
A good paper on issues and solutions is Lumley, Kronmal and Ma, 2006,
Relative Risk Regression in Medical Research: Models, Contrasts, Estimators, and Algorithms
More information about the R-help