Mark,
The theoretical method FE+BE helps, but she needs the procedure
to run this that also works for unbalanced-data. She said that
xtfevd std errors are too small, which is true.
Any suggestions?
Rodrigo.
-------------------------------------------------------------------
Mark wrote:
Rodrigo,
Very clear - I understand now.
How does this help Barbara, though? She says she's interested in the
coefficient on the TI variable in particular. FE+BE doesn't help here,
or am I missing something?
Cheers,
Mark
*
* 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/
