Re: [R] coxme with frailty--variance of random effect?

erse. I.e., theta is chosen first, and the random effects are estimated accordingly. Which is why the variance of the estimated random effects is NOT the same as theta. Did I get this right? Thanks again, Thomas Best wishes, Thomas Chadefaux On Mon, Feb 6, 2012 at 1:56 PM, Terry Therneau wrote: >

[R] coxme with frailty--variance of random effect?

Dear all, This probably stems from my lack of understanding of the model, but I do not understand the variance of the random effect reported in coxme. Consider the following toy example: #--- BEGINNING OF CODE library(su