Dear R users, I have a question on multilevel modeling with R when there is endogenous regressor(s) involved. As far as my economics background concerned, I understand that Hausman-Taylor estimator (via Generalized Least Square) deals with this situation and the package plm does the trick. Unfortunately, plm is created for the conventional 2-level, that is, panel or logitudinal data structure. On the other hand, both lme/lmer (in nlme and lme4 package respectively) are advanced packages for multilevel modeling (via Maximum Likelihood estimators) but I haven't found any option to tackle the potential endogeneity problems. I'm wondering if any one knows a package that handles this type of problem, i.e., multilevel (3-level) with endogenous regresssor. In particular, I'd appreciate it if you can point me to any resource that invokes the HT approach in a multilevel setting. Thanks in advance! Regards, Gerry [[alternative HTML version deleted]]