Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Christopher Baum <kit.baum@bc.edu> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | re:Re: Re: st: Direction of the effect of the cluster command on the |
Date | Thu, 6 Jan 2011 09:52:39 -0500 |
<> Austin wrote Useful to think of super-obs but not quite right. If you have 50 clusters and 100 regressors (with a few thousand obs) but you are only interested in testing one coefficient, you will typically be fine, i.e. you will have negligible bias in the SE thus getting correct inference on average with the CRSE, and it may often be the case that no alternative approach gets you correct inference (except resampling clusters for a cluster-robust bootstrap). So estimating a regression with 50 obs and 100 coefficients is not quite the right analogy--more useful to think of the "effective" sample size as between M (number of clusters) and N (number of obs), computable using "roh" per Kish, L. (1965), Survey Sampling, New York: Wiley (note that the CRSE is also the standard svy estimator). Quite so, Austin; unless you are interested in all the coefficients in a regression, you may not be that concerned about the number of 'super-observations'. The effective sample size is indeed a more useful construct. However it should be noted, for those not that familiar with cluster-robust VCEs, that Stata uses the number of 'super-observations' minus 1 when it reports test statistics. For instance, webuse grunfeld reg invest mvalue kstock time, clu(company) reports an ANOVA F-stat based on 3 and 9 df, where 9 is 10 companies - 1. Likewise, the t-stat pvals are those for 9 df. This is rather important for the original poster of this thread, who was working with 4 clusters (and 3 denom. d.f. in the F, and 3 d.f. in the t). Cheers Kit Kit Baum | Boston College Economics & DIW Berlin | http://ideas.repec.org/e/pba1.html An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html An Introduction to Modern Econometrics Using Stata | http://www.stata-press.com/books/imeus.html * * 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/