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

Re: st: cross validation and student version of STATA "Sophie Barthel"


From   Fred Wolfe <fwolfe@arthritis-research.org>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: cross validation and student version of STATA "Sophie Barthel"
Date   Thu, 27 Feb 2003 14:50:04 -0600

I have a partially written ado file to do 10-fold cross-validation. It is a little crude yet, and is unfinished for bells and whistles, but it works. You may have to hack it a little to do what you want, but send me an email if you would like the file. Cross-validation is most useful when you are "testing" many models.

Fred Wolfe

At 02:09 PM 2/27/2003 -0600, you wrote:

In some contexts, cross-validation means to develop a predictive model in one "training" data set and use it in an independent data set to 'validate' the predictive ability of the model. This reminds me of the split-half coefficient or cronbach's alpha.



I have been asked by a friend to do a bootstrap analysis in STATA for
confidence intervals (which i do know how to do). However, he is also
asking for a cross validation, and to be honest I am a bit stuck as to
what exactly he means by this.
*
*   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/
---------------------------------
Fred Wolfe (fwolfe@arthritis-research.org)
National Data Bank for Rheumatic Diseases
Arthritis Research Center Foundation
Wichita, KS USA

*
*   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