Hi to all,
I am implementing an ivreg estimation.
two stupid questions:
1) I use (in order non to lose too many observation) the first lag of
my independent variable to do my estimation and some second lag....to
apply the test of validity of instruments.
It does not reject the null hypothesis...so the instruments are not
correlated with the error term, is not it?
I thought the the error where ar(1), but if I see this test, it does
not seem. How should I test if teh errors are AR(1)?
2) I try to do the same estimate with ivreg2 using a dummy for any id,
in the way to put them fwl(_const) so to get fixed effects estimates.
In that case it says to me that the matrix is nto of full rank. So I
put also i.id2 in fwl, but then it put an error message, saying that I
did not put i.id2 in the list of regressors...but that it is not true.
Have you got something to suggest me??
Thanks a lot
alessia
*
* 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/
