Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: RE: missing std. errors


From   jverkuilen <jverkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: missing std. errors
Date   Wed, 20 Aug 2008 19:54:32 -0400

"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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index