<>
With only an exogenous variable, the whole "IV" stuff is redundant. You can
use -xtreg- directly... Note that -xtivreg- and -xtreg- throw up the same
results then...
***
use http://fmwww.bc.edu/ec-p/data/macro/abdata.dta
tsset id year
*does not work...
xtivreg2 ys k (n=n), fe
*does work
xtivreg ys k (n=n), fe
xtreg ys k n, fe
***
HTH
Martin
_______________________
----- Original Message -----
From: "Zhongjin Lu" <[email protected]>
To: <[email protected]>
Sent: Friday, April 10, 2009 10:33 PM
Subject: st: single exogenous variable, panel data estimation problem
Dear All:
I have a A simple question with xtivreg2:
I want to run a gmm (re or fe, it doesn't matter to me) regression on
panel data with cross-sectional correlation in error term, that's why
I think the HAC option in Xtivreg2 can solve it.
My question is, I have the following panel data regression:
Y=X*beta+e;
The X is the single and exogenous variable. When I write
the command in STATA, "xtivreg2 y (x=x), fe", it doesn't work! It
seems xtivreg2 need more instruments than endogenous variables? But in
my case, I only have one exogenous variable, what should I do?
Thank you so much.
By the way, it seems, in xtivreg, it doesn't have such requirement.
Since I don't see any formula of xtivreg2, I have no idea about how
does it work.
--
___________________________
Zhongjin Lu (Jack)
Master's Candidate, Department of Economics, Duke University
Alumni, Tsinghua University
Cell Phone +1 919 491 1180
[email protected]
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
