Are there any techniques (built-in Stata or user-provided routines) that allow for Instrumental Variable modeling in a multi-level analysis scenario?
I have health data for counties and states with endogeneity. I have done independent multi-level analyses and IV, but cannot find a routine that allows the use of the two techniques together.
A standard paper is Kim and Frees (2006) "Omitted Variables in Multilevel Modeling", and Andrew Gelman and Jennifer Hill discuss it in Chap.23 of their book and provide a BUGS code example. But, I cannot find a Stata example, discussion, or code reference.
Ken
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
