I have just posted a new version of xtabond2 (2.1.1, dated today) that
fixes a bug that Kit Baum stumbled on. The bug was present in the Mata
code (which matters if you are using Stata 9), in xtabond2 version
2.1.0, released on 12/8/05.
In xtabond2 command lines with more than one gmmstyle() (or gmm())
clause, the bug could cause the program to copy choices from the final
gmm() clause, such as lag limits, over to all other gmm() clauses,
overriding the user's choices.
My apologies for this bug. If you have version 2.1.0 (type "which
xtabond2" to find out), please type "ssc install xtabond2, replace" to
get the update.
--David Roodman
*
* 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/
