Milan Bouchet-Valat <nalimilan at club.fr>
On Tue, 2011-12-20 at 12:00 +0100, r-help-request at r-project.org
wrote:> > Dear all,
> >
> > is there a function similar to extractAIC based on which I can
> extract the
> > BIC (Bayesian Information Criterion) of a coxph model?
> > I found some functions that provide BIC in other packages, but none
> of them
> > seems to work with coxph.
> Use extractAIC(), but set the 'k' argument to log(n), with n the
> number
> of observations in the models (this is explained briefly
> in ?extractAIC).
>
>
For survival models the "effective n" is the number of events rather
than the number of observations. I believe most define BIC in this way.
For a cox model:
> fit <- coxph(Surv(time, status) ~ wt.loss, data=lung)
> fit$nevent
152
Terry Therneau