Hi,

I haven't been able to find a reference for doing this in Stata, but was wondering if anyone has been able to use Lasso for inference with a mixed-effects model, in Stata 16?

In past versions of Stata, I've used melogit with other forms of variable selection methods to fit models with a binary outcome and random intercepts, and have been able to apply Lasso successfully to fit logit models with the new Stata update-- but haven't had much success working through lasso for multilevel/mixed models.

I'd appreciate any advice/direction!