Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: size of cluster robust SEs relative to regular SEs


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: size of cluster robust SEs relative to regular SEs
Date   Thu, 28 Oct 2010 15:15:11 -0400

On Thu, Oct 28, 2010 at 2:34 PM, David Quinn <dxquinnx@gmail.com> wrote:
> Thanks again, Stas.  I fear that this small N problem is really
> trapping me, then.
>
> I do notice that the most significant amount of fluctuation in
> directionality occurs with a suite of dummy variables that were
> constructed out of a categorical variable, wherein I excluded all but
> one of the dummies to prevent perfect collinearity.  Hence, by their
> nature, each of the dummies in the suite contain more zeroes than
> ones.  Not sure if the fluctuation in that regard is natural or not.

If those dummies are concentrated in particular clusters, then yes,
you are in big trouble with them. In the extreme case, if you had
cluster indicators, then you would have zero d.f.s to estimate
variance. If your dummy spans say three clusters, then you have
something like two degrees of freedom for that particular parameters
(and a lousy variance estimator on top of it). If the dummies are
scattered kinda randomly across clusters, then it is not that bad.

-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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   |   Site index