[R] coxph weirdness

2012-07-26 Thread Peter Langfelder
Hi all, I cant' wrap my head around an error from the coxph function (package survival). Here's an example: library(survival) n = 100; set.seed(1); time = rexp(n); event = sample(c(0,1), n, replace = TRUE) covar = data.frame(z = rnorm(n)); model = coxph(Surv(time, event)~ . , data = covar) R

Re: [R] coxph weirdness

2012-07-26 Thread Thomas Lumley
YOu need to update the survival package: from its NEWS file Changes in version 2.36-14 A supposedly cosmetic change to coxph in the last release caused formulas with a . on the right hand side to fail. Fix this and add a case with . to the test suite. -thomas On Thu, Jul 26, 2012 at

Re: [R] coxph weirdness

2012-07-26 Thread Peter Langfelder
On Thu, Jul 26, 2012 at 4:19 PM, Thomas Lumley tlum...@uw.edu wrote: YOu need to update the survival package: from its NEWS file Changes in version 2.36-14 A supposedly cosmetic change to coxph in the last release caused formulas with a . on the right hand side to fail. Fix this and add a