Hi All,

I'm new to this forum, and I hope I can be both of help and find help.

My current problem involves modeling longitudinal data. I have four waves of data and I'm examining employment status and rearrest as endogenous outcome variables. I have built and run a generalized structural equation model (-gsem-) in stata (as opposed to an -sem- model, because my outcomes are generalized, not continuous). All is well with the model, but I can't evaluate the model as a whole. As in something analogous to the ways of evaluating the model fit of an -sem- approach, such as RMSEA or CFI (using -estat gof-). They don't exist currently for -gsem- in Stata 13.

I can look at whether adding or removing variables helps the model using the AIC and BIC (Akaike or Bayesian information criterion) tests. And I can still compare models using likelihood ratio tests. But there's not an overall fit statistic I can report.

So my questions are: (1) If there is not a model fit statistic I can report using -gsem- in Stata, then what should I report when presenting the findings in a report, if anything?; (2) is there another program that claims it does gsem (non-continuous outcomes) that has a way of evaluating the overall model? and (3) anyone familiar with any papers that have used sem with generalized outcomes? I've searched and can't find any published or unpublished. I imagine looking at one and seeing what they report would be very helpful.

Thank you.