Dear Statalisters,
I have two basic questions on specifying a model using
xtabond2.
1. The help file states that "most variables used will
twice in the command line." - implying that not all
have to. Does a specification like:
xi: xtabond2 I a b c i.quarter, gmm(L.I b c)
iv(i.quarter) robust
make sense, where regressor a does not appear in
either the gmm or iv option?
And would the above command be correctly specified in
equation form as follows:
I_it=B_0*I_it-1 + B_1*a_it + B_2*b_it + B_3*c_it+w_t?
2. Most examples that I've looked at include the
dependent variable and/or its lag within the GMM
options. Are there any general guidelines tht should
inform whether or not to specify the DV and/or LDV
within gmm()? Other than some general idea of whether
or not they would be suitable instruments, as well as
looking at the Hansen stat? Is it the "usual" practice
to include them?
I'd be grateful for any suggestions on any part(s) of
what I've asked above.
Best,
Crystal Lopez
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
*
* 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/
