I actually had this same problem and wrote the author of the user-written
command for help. As it turned out, one of my variables was problematic in that
it had too little variation. I excluded this variable from the regression and
also inputed the command -drop if missing(variable name here) | missing(
variable2 name here).... This helped and I was able to run it.
I do have another question however. Does anyone know how to cluster with this
command and/or if there is an analogous command to 'overid' for
overidentification testing purposes that will work with this command?
April M. Knill
Consultant, Finance Research Group
The World Bank Group
Mail Stop MC3-300
1818 H Street, NW
Washington, DC 20433
202.458.8699
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/