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

st: RE: Senstitivity


From   "Steichen, Thomas" <[email protected]>
To   <[email protected]>
Subject   st: RE: Senstitivity
Date   Wed, 18 Sep 2002 10:47:44 -0400

JT Sheu writes:

> I am not sure whether my question is suitable in this 
> discussion group. I am performing sensitivity and specificity 
> test using STATA command of -diagtest. Beside to values of 
> sensitivity and specificity, I also obtain NPV(negative 
> predicted value) and PPV(positive predicted value) from this 
> command as well. I am just wondering whether these two 
> values(NPV and PPV) are common to report in paper or not?

JT... 

First, do NOT use -diagtest-.  Some of the confidence 
intervals in it are calculated via an incorrect formula. 
Instead, use the replacement program -diagt- and its companion 
immediate program -diagti-.  These programs, by Paul Seed and 
Aurelio Tobias (the author of -diagtest-), offer additional
features beyond -diagtest-.

Second, Paul Seed has another release of these programs under
development that should be ready (I suspect the release is  
delayed due to the write-up, not the code).  You may wish to
contact him for details.

Third, as to your question... I've seen them published, but 
how commonly reported, I don't know.  I think it is more
important for you to consider whether these values help you
communicate knowledge to your readers.  If so, do so!  Most
of the statistics reported are related, so you can pick a
subset that best explains details in your data.  If you think
it is important to report the proportion of correct results
(when split into positive predictions and negative predictions),
then do so.  If you think they add nothing to your report,
then do not report the values.

Tom
CONFIDENTIALITY NOTE:  This e-mail message, including any attachment(s), contains information that may be confidential, protected by the attorney-client or other legal privileges, and/or proprietary non-public information.  If you are not an intended recipient of this message or an authorized assistant to an intended recipient, please notify the sender by replying to this message and then delete it from your system.  Use, dissemination, distribution, or reproduction of this message and/or any of its attachments (if any) by unintended recipients is not authorized and may be unlawful.




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