Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "JVerkuilen (Gmail)" <jvverkuilen@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: clustered SE smaller than |
Date | Thu, 21 Mar 2013 12:10:05 -0400 |
On Thu, Mar 21, 2013 at 11:22 AM, Robert G. LaChausse <rlachaus@csusb.edu> wrote: > > My understanding is that the robust SE should be larger than the regular OLS > SE. I found that the robust SE were actually a little smaller (ie .17 > regular SE and .14 robust SE). The cluster sizes range from 13 to 321. There > are 22 clusters. That's typically what happens but it's by no stretch of the imagination required by robust standard errors. However I do worry a bit about the hugely unbalanced cluster sizes, with the relatively small number of clusters. That can create trouble. You might want to try a clustered bootstrapping approach, which Stata does very easily, because that depends much less on asymptotics. Just make sure to boost the number of replications up to a larger value (say 999). * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/