Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: collinearity - new versions of coldiag2 and perturb


From   John Hendrickx <[email protected]>
To   statalist <[email protected]>
Subject   st: collinearity - new versions of coldiag2 and perturb
Date   Fri, 10 Dec 2004 09:51:00 -0800 (PST)

Thanks to Kit, new versions of -coldiag2- and -perturb- are available
from SSC. -coldiag2- is for calculating condition indexes and
variance decomposition proportions in order to evaluate collinearity
and track its sources. The new version of can be use as a
post-estimation command like -vif-. There's a "fuzz" option to
suppress printing small variance decomposition values and it's also
more intelligent about missing values.

-perturb- is a general tool for evaluating collinearity and
ill-conditioning. It adds random noise to selected variables, then
re-estimates the model a number of times to determine parameter
stability. If small changes to variables cause disproportionate
changes to parameters, then the model is ill-conditioned. 

-perturb- works with categorical variables. In that case, the
variables are reclassified according to table of reclassification
probabilities. The new version lets you simply specify a degree of
misclassification, then determines suitable reclassification
probabilities that approximate the desired degree of
misclassification such that these have an "appropriate" pattern of
association. Comments invited!

John Hendrickx


		
__________________________________ 
Do you Yahoo!? 
All your favorites on one personal page � Try My Yahoo!
http://my.yahoo.com 
*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index