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

Re: st: diagt and accuracy

From   Roger Newson <[email protected]>
To   [email protected]
Subject   Re: st: diagt and accuracy
Date   Fri, 10 Jun 2005 18:06:35 +0100

At 17:23 10/06/2005, Ronan Conroy wrote:
[email protected] wrote:

Originally I wanted to have ROC area (as calculated in -diagt-), but I was
told that with a dichotomous test ROC area wasn't relevant, ie, I didn't
have different cutpoints at which to operate the test. That's when she
mentioned accuracy defined with prevalence.

In fact, ROC area doesn't account for prevalence either...
The ROC area is still useful, because it is the Mann Whitney statistic - the probability that if a person with and a person without the condition are taken at random, the person with the condition will have the higher score. In clinical trials, it becomes the probability that a person on the superior treatment will have a superior response. Both are very useful clinical information.
For more about the ROC area, confidence intervals for the difference between 2 ROC areas for the same disease in the same population, and the connection with the Mann-Whitney ranksum test, see Newson (2002). This can be downloaded from my website (see my signature below), either using a browser or from within Stata by typing

findit params

and clicking on the hypertext for the paper.

I hope this helps.



Newson R. 2002. Parameters behind "nonparametric" statistics: Kendall's tau, Somers' D and median differences. The Stata Journal 2(1): 45-64.

Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
Division of Asthma, Allergy and Lung Biology
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: [email protected]

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

* For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index