Hi there,
and how about this:
nsim<-500
est<-NULL
for(i in 1:nsim){
fit <- geese(x ~ trt, id=subject, data=data_gee, family=binomial,
corstr="exch", scale.fix=TRUE)
.............
corr_gee<-summary(fit)$correlation[1]
se_corrgee<-summary(fit)$correlation[2]
est<-data.frame(rbind(est,cbind(corr_gee, se_corrgee)))
}
or may be if you change c() by cbind() may also work
on your example.
bests
milton
On Wed, Sep 2, 2009 at 2:20 PM, Seunghee Baek
<seunghee@mail.med.upenn.edu>wrote:
> Hi R-users,
> I have a problem for updating the estimates of correlation coefficient in
> simulation loop.
> I want to get the matrix of correlation coefficients (matrix, name: est)
> from geese by using loop(500 times) .
> I used following code to update,
>
> nsim<-500
> est<-matrix(ncol=2, nrow=nsim)
> for(i in 1:nsim){
>
> fit <- geese(x ~ trt, id=subject, data=data_gee, family=binomial,
> corstr="exch", scale.fix=TRUE)
> .............
> corr_gee<-summary(fit)$correlation[1]
> se_corrgee<-summary(fit)$correlation[2]
> est[i,]<-c(corr_gee, se_corrgee)
> }
>
> But, I got an error message : number of subsript in matrix does not match.
>
> Can somebody help me? I think sumary(fit)$correlation[1] print out the name
> of estimates.
> I guess it might cause problem to be updated in matrix.
> I want to know other way to print out the estimates of correlation if it
> helps.
>
> Thanks,
> Becky
>
> ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
>
http://www.R-project.org/posting-guide.html<http://www.r-project.org/posting-guide.html>
> and provide commented, minimal, self-contained, reproducible code.
>
[[alternative HTML version deleted]]