Re: [R] correlated data

2007-12-07 Thread Prof Brian Ripley
Look at package lme4. If you have code set up for glme(), adapting it to glmmPQL in package MASS will be rather easier, but if you are starting afresh I would suggest you learn the lme4 syntax straight away. On Fri, 7 Dec 2007, HAKAN DEMIRTAS wrote: > Thanks for your reply. I am looking for an

Re: [R] correlated data

2007-12-07 Thread HAKAN DEMIRTAS
Hi, Thanks for your reply. I am looking for an R package that has the same functionality of the glme() function within the correlatedData library in SPlus 6.2 or above. This generic function fits a generalized linear mixed-effects model in the formulation described in Breslow and Clayton (19

Re: [R] correlated data

2007-12-07 Thread Uwe Ligges
HAKAN DEMIRTAS wrote: > Hi, > > Is there an R library that has the same functionalities of Splus7.0+ library > correlatedData? If you mean an R *package* and tell us what functionalities are available in "Splus7.0+ library correlatedData", we might be able to help. Uwe Ligges > I'd appreci

Re: [R] correlated data

2007-10-17 Thread Sarah Goslee
Hi, What's "properly"? What are you trying to do - you don't ever say. I think you need to set population=TRUE to get the result you appear to want (sorry, that's not entirely clear in the docs, I see). > x <- 1:100 > y <- corgen(x=x, r=.8, epsilon=.01, population=TRUE)$y > cor(x, y) [