I am attempting to run a parallel process LGM with 497 cases. One variable is Poisson distributed so I use the COUNT option. The other variable is normally distributed. We have four timepoints for each variable. When I run the LGMs of the two variables separately, they converge nicely and I have two good-fitting models. When I try to put them together in a parallel process model, nothing I do seems to help them converge. Is the problem that you simply cannot have one count and one continuous variable in a PPLGM? I couldn't find any information in the manual, so thought I would turn to the discussion board. Thanks for any help you might be able to provide!