Is it possible to use Penalized Quasi-Likelihood estimation for GLMM model ? I found the meglm glm and related command all uses Lap or AGQH for estimation, but no one contained the PQL, is there any method to use it ? Thanks