Statalist


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

st: SDT/ROC equal variance test


From   James <james.e.heyman@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: SDT/ROC equal variance test
Date   Thu, 23 Apr 2009 10:31:50 -0500

I'm working on an SDT problem and the key hypotheses I'm testing requires the equal-variance assumption to hold. Is there a way to test this with the roc functions? I assume there is and am justing missing it. What I have so far:

1. roctab produces a nice monotonic convex ROC curve. But this is a non-parametric test so I'm not sure that helps much.

2. the slope coefficient in rocfit is .97. Is that the slope of the z-transformed ROC, which presumably would mean I'm good to go. Does the fact that the ROC areas are similar across the non-parametric & binormal models do anything for me?

Thanks much.

James

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



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index