Predictive performance of mixclogit

[Thang Nguyen]

Dear all,

I would like to assess in-sample and out-of-sample predictive performance of mixclogit.

Could you instruct me to do this?

Thank you.

Best regards.

[Erik Ruzek]

I am not familiar with mixclogit. There is a program called mixlogit, first published in the Stata Journal.

At any rate, with mixed effect models of any variety, you need to be clear about whether you are making predictions about clusters or about units within clusters (observations). The former cannot be predicted while the latter can. See this thread.