Hi,
I just wanted to close my query on missing std. errors. The suggestions that Sebastian and jverkuilen provided were helpful. Only a particular dummy/indicator variable (of the 50) was creating trouble!
Nishant
--- On Wed, 8/20/08, jverkuilen <[email protected]> wrote:
> From: jverkuilen <[email protected]>
> Subject: RE: st: RE: missing std. errors
> To: [email protected]
> Date: Wednesday, August 20, 2008, 7:54 PM
> "Forshee, Richard" wrote:
>
> >Have you excluded a reference category? If not, your
> dummy variables
> will be perfectly collinear with the constant. >
>
> Stata should handle this though often in an arbitrary
> way---fitting as the choice is mathematically (but not
> substantively) arbitrary.
>
>
> Nishant Dass wrote:
>
> >First of all, I am using Stata/SE 10.0 on Windows.>
>
> -update- Stata 10.1 is available.
>
>
> I can't tell from your output (the output doesn't
> have the code) but it looks like you are wanting to run
> clustered robust SE and also have a lot of dummies which
> probably correspond to the clusters. Is that right? If so, I
> am not surprised the parameter estimate covariance matrix
> which you need for the standard errors is singular---you
> don't deserve an estimate given your model, which is
> probably misspecified for cluster SEs.
>
> I wonder if what you really want is -xtreg- with fixed
> effects?
>
>
>
>
>
> *
> * 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/
