Estimate SE for odds ratios
NWPope opened this issue · comments
NWPope commented
It looks like lme4::ranef calculates the correct standard errors for the random effects on the odds scale when specifying condVar = TRUE, transf = "exp". The object returned by ranef has the conditional variances stored as an attribute, which when passed to lattice::dotplot plot the CI's very nicely. Would it be possible to do something similar with REsim to produce the SEs on the odds scale?
NWPope commented
After testing this it doesn't seem to work. The transf is an argument to lattice::dotplot that is no longer recognized