[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: ROC curves for ordinal outcomes

From   "Newson, Roger B" <>
To   <>
Subject   st: RE: ROC curves for ordinal outcomes
Date   Thu, 12 Feb 2009 21:42:45 -0000

The ROC area is a special case of Harrell's c index, which still exists
for non-binary ordinal outcomes. Confidence intervals for Harrell' c
indices and their differences can be calculated using the -somersd-
package, downloadable from SSC using the -ssc- command in Stata. The
on-line help contains some references, some of which can be downloaded
from the Stata Journal website at
and others of which can be downloaded from my website (see my signature

I hope this helps.

Best wishes


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Web page:
Departmental Web page:

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

-----Original Message-----
[] On Behalf Of
Sent: 12 February 2009 21:30
Subject: st: ROC curves for ordinal outcomes

  I'm trying to compare 3 screening tests. My out come variable is not
binary. My outcome variable is ordinal 
with 8 categories. Is there a way in STATA to produce a ROC curve to
compare these tests for ordinal outcomes?

Or is it possible to do a ordinal regression and show the results as ROC

I would really appreciate your reply.


*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index