Is there a simple way to extract the frailty variance parameter (theta)?
Below is an example of the method that I have.
library(survival)
data(rats)
rfit2b <- survreg(Surv(time, status) ~ rx +
frailty.gaussian(litter, df=13, sparse=TRUE), rats )
{
#According to rfit2b$printfun, this is the way to extract "theta"
if (!is.null(rfit2b$history[[1]]$history))
theta <- rfit2b$history[[1]]$history[nrow(rfit2b$history[[1]]$history), 1]
else theta <- rfit2b$history[[1]]$theta
}
theta # This is 0.01926811
#It would be a nice feature if the final variance estimate
#was directly accessible by:
rfit2b$history[[1]]$theta # This is 0.01926953 != theta
Could we make it that way?
Sincerely,
Jerome Asselin
P.S.: I am using the survival package 2.9-6 on R 1.6.2 on RedHat Linux 7.2.
--
Jerome Asselin (J?r?me), Statistical Analyst
British Columbia Centre for Excellence in HIV/AIDS
St. Paul's Hospital, 608 - 1081 Burrard Street
Vancouver, British Columbia, CANADA V6Z 1Y6
Email: jerome at hivnet.ubc.ca
Phone: 604 806-9112 Fax: 604 806-9044