Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: detecting a dropped variable programatically


From   "David Harrison" <david.harrison@icnarc.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: detecting a dropped variable programatically
Date   Fri, 19 May 2006 09:11:29 +0100

e(df_m) will tell you how many degrees of freedom your model has. If
this is less than the number of predictors (the columns of e(b) minus 1)
then something has been dropped.

David

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Mark
Sent: 19 May 2006 08:55
To: statalist@hsphsun2.harvard.edu
Subject: st: detecting a dropped variable programatically

I am writing a program that runs through potentially tens of thousands
of regressions.  During this evaluation, I sometimes end up with
collinear variables.  (Not really, but very nearly collinear, and so
numerically stata ends up thinking they are.)  I know that I can look
thru the e(V) matrix looking for a value of zero on the diagonal.  But
doing this for every regression means another loop inside the main loop
that may run tens of thousands of times.

Does anyone know of a way to detect a dropped variable without a loop?

What would be nice is if the regression routine would return an
indicator of a dropped variable.  But I dont see this, nor any obvious
way to find it based on the return values except for searching thru
e(V).  Any ideas appreciated.

Thanks!

Mark

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index