I am running a bivariate sex-limitation model with an age variable as a moderator on both means and variances/covariances. All models (full and nested) seem to work pretty well, estimates are reasonable and there is no error while converging (not even code GREEN). However, when I look at the standard erros, I can see that some of them turn out to be NAN's. What could be a reason of it and how can I fix it?
Thank you beforehand.