Ada,
It's not a glitch, it's by design. Have a look at the discussion of
robust variance calculation in the -regress- manual entry.
--Mark
Date sent: Fri, 24 Sep 2004 15:14:05 +0100
From: Ada Ma <[email protected]>
Organization: Department of Economics, University of Aberdeen
To: [email protected]
Subject: st: a glitch in the reg, cluster option?
Send reply to: [email protected]
Hello Statalist,
I have been doing linear regressions using the _cluster_ option.
The first few lines of results would show something like:
**********************************************************************
*** Regression with robust standard errors Number of obs = 4247
F(130, 3413) = 10.50
Prob > F = 0.0000
R-squared = 0.2319
Number of clusters (personid) = 3414 Root MSE = .47171
**********************************************************************
***
Now you can see that Stata says there are 3414 unique individuals in
the data set. I happened to have counted the number of individuals
and found it to be different (my numbers are always smaller). I found
that the number of clusters/individuals specified in the Stata results
to be the (degrees of freedom + 1) for all the regressions I have
performed.
Could someone tell me whether I have spotted something strange about
the cluster command or am I missing something?
Cheers,
Ada Ma
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
Prof. Mark E. Schaffer
Director
Centre for Economic Reform and Transformation
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS UK
44-131-451-3494 direct
44-131-451-3008 fax
44-131-451-3485 CERT administrator
http://www.som.hw.ac.uk/cert
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
