Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

RE: st: Trouble with estat classification


From   "Dukes, Jonathan" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   RE: st: Trouble with estat classification
Date   Wed, 5 Feb 2014 19:37:39 +0000

I should preface with the fact that I used a dichotomized variable in the roctab (i.e. 0=Maternal Age LT 35, 1=Maternal Age GT 35).

roctab USabnormal matage, summary detail
------------------------------------------------------------------------------
                                           Correctly
Cutpoint      Sensitivity   Specificity   Classified          LR+          LR-
------------------------------------------------------------------------------
( >= 0 )          100.00%         0.00%       30.38%             1.0000
( >= 1 )           12.50%        80.00%       59.49%            0.6250       1.0938
( >  1 )            0.00%       100.00%       69.62%                                 1.0000
------------------------------------------------------------------------------

logistic USabnormal matage
estat classification

              -------- True --------
Classified |         D            ~D  |      Total
-----------+--------------------------+-----------
     +          |         0                  0  |          0
     -           |        24              55  |         79
-----------+--------------------------+-----------
   Total     |        24              55  |         79

Classified + if predicted Pr(D) >= .5
True D defined as USabnormal != 0
--------------------------------------------------
Sensitivity                     Pr( +| D)    0.00%
Specificity                     Pr( -|~D)  100.00%
Positive predictive value       Pr( D| +)       .%
Negative predictive value       Pr(~D| -)   69.62%
--------------------------------------------------
False + rate for true ~D        Pr( +|~D)    0.00%
False - rate for true D         Pr( -| D)  100.00%
False + rate for classified +   Pr(~D| +)       .%
False - rate for classified -   Pr( D| -)   30.38%
--------------------------------------------------
Correctly classified                        69.62%
--------------------------------------------------



Respectfully,


Jonathan



-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Richard Goldstein
Sent: Wednesday, February 05, 2014 1:27 PM
To: [email protected]
Subject: Re: st: Trouble with estat classification

there is nothing wrong here - they are just doing different things:
estat classification uses one cutpoint; -roctab- uses more than one as shown in the documentation

Rich

On 2/5/14, 2:17 PM, Dukes, Jonathan wrote:
> Dear Statalist,
>
> I used estat classification after a logistic regression to find the
> sensitivity,
specificity, and positive predictive value.  Using estat classification I get a sensitivity of 0% and a specificity of 100%. However, using roctab I get a sensitivity of 12.5% and specificity of 80%. Does anybody have an idea of what is wrong with the estat classification statement? Thank you for your help.
>
>
>
>
> Very Respectfully,
>
> Jonathan
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

________________________________

The materials in this email are private and may contain Protected Health Information. If you are not the intended recipient be advised that any unauthorized use, disclosure, copying, distribution or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via return email.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index