try including your dummy using "diffvars(varlist)", which allows you to tell stata not to difference the specified varlist (by default xtabond take first differences of your data as the relevant variables for the estimation procedure). For the instruments specification try maxldep or maxlags
I hope this helps
Marco
-----Mensaje original-----
De: Vidya Mahambare [mailto:[email protected]]
Enviado el: Lunes, 03 de Noviembre de 2003 10:25 a.m.
Para: [email protected]
Asunto: st: xtabond
I am a new user of STATA. I am trying to run a dynamic panel data model for
cobb-douglas production function using xtabond command. I have year dummies in
the model as well. When I run the model, first three year dummies are dropped
due to collinearity. Could anyone explain why should this happen? I was
expecting only one dummy variable to be dropped. Also, is there a way to specify
lags of variables, say t-2 and and not t-1, for instruments?
thanks
vidya
thanks
Vidya
*
* 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/
*
* 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/
