Your categorical variable "onefive" appears to take only 2 values - making fitting the binormal model impossible.
David
-----Original Message-----
From: b. water [mailto:[email protected]]
Sent: 17 February 2005 15:36
To: [email protected]
Subject: st: error in rocfit?
dear listers,
i just ran rocfit on a dataset obtaining the following output. surely the
std error or the AUC can't be right (>1). would appreciate advice and
comments. fwiw, roctab showed standard error of 0.0724.
Binormal model of fbs on onefive                  Number of obs   =
39
Goodness-of-fit chi2(-1) =        0.00
Prob > chi2             =           .
Log likelihood          =  -23.115784
------------------------------------------------------------------------------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
   intercept |   1.085096  4.579e+03     0.00   1.000   -8.973e+03
8.975e+03
   slope (*) |   0.912109  5.161e+03    -0.00   1.000   -1.011e+04
1.012e+04
-------------+----------------------------------------------------------------
       _cut1 |   0.887147   0.362530     2.45   0.014     0.176601
1.597692
------------------------------------------------------------------------------
------------------------------------------------------------------------------
             |                    Indices from binormal fit
       Index |   Estimate    Std. Err.                    [95% Conf.
Interval]
-------------+----------------------------------------------------------------
    ROC area |   0.788637 382.665741                    -7.492e+02
750.799708
    delta(m) |   1.189655  1.712e+03                    -3.354e+03
3.356e+03
        d(e) |   1.134972  1.726e+03                    -3.381e+03
3.383e+03
        d(a) |   1.133775  1.871e+03                    -3.665e+03
3.668e+03
------------------------------------------------------------------------------
(*) z test for slope==1
thank you,
bw
_________________________________________________________________
Don't just search. Find. Check out the new MSN Search!
http://search.msn.click-url.com/go/onm00200636ave/direct/01/
*
*   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/
*
*   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/