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

st: Evaluating a prediction model - a general statistical question


From   "Peter Jepsen" <peterjepsen@privat.dk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Evaluating a prediction model - a general statistical question
Date   Tue, 10 May 2005 00:50:15 +0200

Dear Statalisters,

I am examining whether a liver cirrhosis patient's galactose elimination
capacity predicts 1-year survival probability. My study includes around
1,000 patients. This is my first shot at creating a prediction model based
on logistic regression. I understand that it is advisable to draw a, say,
two-thirds sample and use this sample to estimate the parameters and then
test the model on the last third. Now, I have read that there is a way to
improve this method by doing it a number of times with random two-thirds. I
have searched for more information on this, but I simply don't know where to
start. Could somebody please tell me where to look (books, articles, Stata
commands)?

Thank you in advance.

Peter.

*
*   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