Did you check that the set of new_attid effects was complete? I.e.
was one of the fixed effects dropped instead of pitt being dropped?
There should be 1092 coefficients listed: _Inew_attid_2-1093. Is there
variation in pitt across time within each panel?
It is possible that collinearity can produce a dropped variable in one
command and not another, since Stata checks for collinearity by
checking for full rank conditions within certain tolerances, and the
matrices checked can be different in different commands, but I would
expect the standard errors to look very suspicious in Model 2 in that
case, and the SE on pitt (.1575) does not look near infinity at all in
Model 2.
On 3/12/06, Scott Hankins <[email protected]> wrote:
> The 1st uses "xtreg" to estimate a fixed effects model (along with "xi" for the
> year effects). The 2nd uses "regress" and "xi" to generate dummy variables for
> the individual fixed effects as well as year effects.
>
> The only difference between the two routines is the coefficient on pitt is
> "dropped" when I use xtreg. I thought this was a multi-colinearity problem until
> I ran the regression with "regress" and dummies. It may still be a
> multi-colinearity problem as I do not believe the estimated coefficient is correct.
>
> ********* 2nd model (note: I removed the new_attid effects) *********
*
* 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/
