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

st: what's wrong with ROC curves


From   BISSERY Alvine <Alvine.BISSERY@hop.egp.ap-hop-paris.fr>
To   statalist@hsphsun2.harvard.edu
Subject   st: what's wrong with ROC curves
Date   Mon, 15 Sep 2003 11:03:30 +0200

Dear all, 

I have a problem with the roctab function.
I have 2 variables : "groupe " : 1 for disease, 0 for no disease
		        "outcome" wich is a continous variable measuring
PTH, with values from 12 to 58

- roctab groupe outcome, d - gives :

Detailed report of Sensitivity and Specificity
----------------------------------------------------------------------------
--
                                           Correctly
Cut point     Sensitivity   Specificity   Classified          LR+
LR-
----------------------------------------------------------------------------
--
( >= 12 )           0.00%       100.00%       60.00%
1.0000
( >= 14 )           8.33%       100.00%       63.33%
0.9167
( >= 25 )           8.33%        94.44%       60.00%       1.5000
0.9706
( >= 29 )           8.33%        72.22%       46.67%       0.3000
1.2692
( >= 33 )           8.33%        66.67%       43.33%       0.2500
1.3750
( >= 36 )           8.33%        61.11%       40.00%       0.2143
1.5000
( >= 38 )           8.33%        50.00%       33.33%       0.1667
1.8333
( >= 40 )           8.33%        38.89%       26.67%       0.1364
2.3571
( >= 42 )           8.33%        33.33%       23.33%       0.1250
2.7500
( >= 44 )           8.33%        27.78%       20.00%       0.1154
3.3000
( >= 45 )          16.67%        11.11%       13.33%       0.1875
7.5000
( >= 46 )          16.67%         0.00%        6.67%       0.1667     
( >= 47 )          25.00%         0.00%       10.00%       0.2500     
( >= 48 )          33.33%         0.00%       13.33%       0.3333     
( >= 50 )          41.67%         0.00%       16.67%       0.4167     
( >= 52 )          50.00%         0.00%       20.00%       0.5000     
( >= 53 )          66.67%         0.00%       26.67%       0.6667     
( >= 54 )          75.00%         0.00%       30.00%       0.7500     
( >= 55 )          83.33%         0.00%       33.33%       0.8333     
( >= 58 )          91.67%         0.00%       36.67%       0.9167     
( >  58 )         100.00%         0.00%       40.00%       1.0000     
----------------------------------------------------------------------------
--

How is it possible to have a specificity = 100% and sensitivity = 0% for the
first cut point ? all people >= 12 have to be classifed has "disease people"
because test is abnormal and so sensivity must be 100% and specificity = 0 %

Can someone help me ?!!! 


Alvine Bissery



***************************************************************
Alvine BISSERY
Biostatisticienne

Hôpital Européen Georges Pompidou
Centre d' Investigations Cliniques
20-40, rue Leblanc 
75908 PARIS Cedex 15
France

tél: +33 1 56 09 29 13
fax: +33 1 56 09 29 29 

e-mail: alvine.bissery@hop.egp.ap-hop-paris.fr


*
*   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 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index