Statalist


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

RE: st: DW stats for pooled data


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   stata list <statalist@hsphsun2.harvard.edu>
Subject   RE: st: DW stats for pooled data
Date   Mon, 14 Dec 2009 14:13:23 -0800 (PST)

-- Anita Sayal wrote:
> Could someone please tell how does one detect autocorrelation problem
> in pooled observations.

Stata will automatically detect (perfect) autocorrelation. You will see
that as it drops one of the variables. As it is perfect autocorrelation,
it does not matter which one; the variables are indistinguishable anyhow.
There are people who worry about high but non-perfect correlation between
covariates. If you are one of those, you can look at it by displaying a 
correlation matrix with -corr-. However, it is good to remember that such 
non-perfect correlations are not a problem, the only possible concern 
could be that such variables will eat a lot of statistical power if the
correlations are extremely high, but that is exactly as it should be. 

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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