On Fri, 2002-12-06 at 14:52, Cleves, Mario A wrote:
> Arnold Kester <mailto:[email protected]> asked:
>
> >I just produced the weirdest ROC curve I've ever seen, and I suspect there's a bug in roctab. >The curve jumps up and down which is by definition not possible. For your amusement and
> >curiosity I append the data. I am aware that as a diagnostic, x is worthless to predict y.
>
> I suspect that Arnold has one of several pathologic data cases that are better analyzed using Lorenz-like curves as proposed by Lee(1999), and referred to on page 138 of the Q-St Reference manual.
>
I don't agree, it's not my point whether the analysis is proper for
these data. Even when the data are 'pathological', which I doubt, the
ROC curve should be drawn as it is defined. And by definition, the curve
cannot be non-monotone. If the curve is drawn wrong in this case, how
can we be confident that it will be correct in other instances?
Regards,
Arnold Kester
*
* 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/
