Cox + bootstrap

["Heymans, MW" <mw.heymans@vumc.nl>]

Hi all,

 

Help is appreciated on this one.

I wrote some simple lines to apply the bootstrap and Cox regression (used Canty's and Harrells functions).

 

bw.fun <- function(data, i){

   d <- data[, i]

   t <- d$dag2

   e <- d$st2

   dd <- datadist(d = d)

   options(datadist = 'dd')

   Srv <- Surv(t,e)

 

   fit.bw <- cph(form.m, data = "" surv=TRUE, time.inc = 180, x = T, y = T)

   fb <- fastbw(fit.bw)

   list(coef(fb))

}

 

bw.boot <- boot(data = "" statistic = bw.fun, R = 5)

 

However, the regr. coeffiecients are not stored in bw.boot$t and I do not know why? (it is the same with other fastbw values (like names.kept, factors.kept). It is strange because If I do fastbw on the original sample about 5 factors are kept in the model. Can somebody help me with the solution?

 

Thank!

 

Martijn

 

PS. I also tried other selection levels for the variables, but no difference (i.e. no selections).