Yuval Sapir said the following on 10/12/2006 1:08 PM: > Hi, > I'm trying to perform a glm model on count data (poisson distribution of > the errors) where data are nested. > glmmodel<-glm(y~x/z,poisson) > x and z are factors, z nested within x, y is count data. > In that point the R just "stuck" and not respond anymore. I tried > glmmodel<-glm(y~x,poisson) > and there were no problems. Trying > glmmodel<-glm(y~x/z) > gave the same no response. Similar problem with continuous data (normal > distribution of the errors). > I am using R 2.3.1 on Windows XP > Thanks > Yuval >
Please see the signature "and provide commented, minimal, self-contained, reproducible code." Namely, what is 'x' and 'z'? My guess is ~x/z produces so large a model matrix you are approaching your limits of memory. But, there's no way to tell without more information. HTH, --sundar ______________________________________________ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.