Skip to content
Prev 6333 / 20628 Next

Specifying a (simple?) two level model

I would focus on merging clusters. A smaller number of cluster will not only solve the problem om complete separation. You will get a smaller variance-covariance matrix for the random effects. That will speed up the fit as well.
Something like this

library(reshape)
cast(cluster ~ country, data = your.data.frame, value = "your.response", fun = mean)
sum(cast(cluster ~ country, data = your.data.frame, value = "your.response", fun = mean) %in% c(0, 1), na.rm = TRUE)