Re: st: re: obtaining overlapping ROC graphs

[Roger Newson <roger.newson@kcl.ac.uk>]

At 12:31 27/08/02 -0400, Devendra Amre wrote:
> Hi,
>
> I was wondering if it is possible to plot 2 overlapping graphs (not 2 
> graphs on the same page as is possible with the statagraph combine 
> command) each representating different regression equations (example one 
> in sample another out-of-sample). I wish to compare 2 such ROC curves.
Devendra does not state how the data are formatted. However, if the false 
positive rate is -fpr-, and the true positive rate (ie sensitivity) is 
-tpr-, and the indicator for being in-sample is -insamp- (equal to 1 if 
in-sample, 0 if out of sample), then Devendra might write

gene tpr0=tpr if insamp==0
gene tpr1=tpr if insamp==1
graph tpr0 tpr1 fpr,s(oo) c(LL)

and the graph would then have 2 lines, 1 for -tpr1- against -fpr- and one 
for -tpr0- against -fpr-. The general principle is that Stata likes to 
graph multiple Y-variables against a single X-variable. If there are 
multiple groups, and you want Stata to draw multiple lines (one per group), 
then Stata likes you to create multiple Y-variables (one per group), and 
plot them against the X-variable.

I hope this helps.

Roger


--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
King's College London
5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom

Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
  or 020 7848 6605 International +44 20 7848 6605
Email: roger.newson@kcl.ac.uk

Opinions expressed are those of the author, not the institution.

*
*   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/