R help - Dec 2013 - combine glmnet and coxph (and survfit) with strata()

Dear All,

I want to generate survival curve with cox model but I want to estimate the
coefficients using glmnet. However, I also want to include a strata() term
in the model. Could anyone please tell me how to have this strata() effect
in the model in glmnet? I tried converting a formula with strata() to a
design matrix and feeding to glmnet, but glmnet just treats the strata()
term with one independent variable...

I know that if there is no such strata(), I can estimate coefficients from
glmnet and use "...init=selectedBeta,iter=0)" in the coxph. Please
advise
me or also correct me if I'm wrong.

Thank you very much!

Best,

Jieyue

	[[alternative HTML version deleted]]

I'm also curious how to use glmnet with survfit -- specifically, for use
with interval regression (which, under the hood, is implemented using
survfit).  Can you show how you converted your Surv object formula to a
design matrix for use with glmnet?

Thanks,
-Aaron


On Sun, Dec 8, 2013 at 12:45 AM, Jieyue Li <jieyueli82@gmail.com> wrote:
> Dear All,
>
> I want to generate survival curve with cox model but I want to estimate the
> coefficients using glmnet. However, I also want to include a strata() term
> in the model. Could anyone please tell me how to have this strata() effect
> in the model in glmnet? I tried converting a formula with strata() to a
> design matrix and feeding to glmnet, but glmnet just treats the strata()
> term with one independent variable...
>
> I know that if there is no such strata(), I can estimate coefficients from
> glmnet and use "...init=selectedBeta,iter=0)" in the coxph.
Please advise
> me or also correct me if I'm wrong.
>
> Thank you very much!
>
> Best,
>
> Jieyue
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]