R package (relaxed lasso for Cox's proportional hazards model) Is there any R package to implement relaxed lasso for Cox's proportional hazards model? Thank you.
 A: I found this package after some googling around. I have never used it so can't speak to its efficacy, but from the description I think it's what you're looking for:

We implement a cocktail algorithm, a good mixture of coordinate decent, the majorization-minimization principle and the strong rule, for computing the solution paths of the elastic net penalized Cox’s proportional hazards model.



*

*http://code.google.com/p/fastcox/

*http://cran.r-project.org/web/packages/fastcox/
The package references the following paper, which appears to have been accepted into a journal but has not yet been published:


*

*Yang, Y. and Zou, H. (2012), A Cocktail Algorithm for Solving The Elastic Net Penalized Cox’s Regression in High Dimensions. Statistics and Its Interface (accepted).

A: Did you ever find the answer to your question ???
I have been using the cv.glmnet function to identify the most informative set of features.
I then generated a regular coxph with these features and sort of got an estimate for the HRs.
However, this is not an optimal solution as the coxph model doesn't include the penalties and is therefore not as accurate.
