I think that this can be done in Stata using -simulate-. Samples of various
sizes with given areas under the reciever operating characteristic curves could
be created with -drawnorm outcome predictor, corr(1 `rho' \ `rho' 1)
n(`sample_size')-. Then -replace outcome = outcome > 0.5-, and -replace- the
predictor according to the assumption for its distribution. (For a normally
distributed predictor, AUC given by -lroc- after -probit outcome predictor- is
(`rho' + 1) / 2). Create two such samples (one for each of the two indepedent-
sample comparison groups) with their given AUCs determined by their respective
`rho' values used in -drawnorm-, and -append- them, using a grouping indicator
variable to tell them apart and to reference in the -by- option in -roccomp-.
Power to discriminated the two groups at each sample size and level of Type I
error rate can then be estimated based upon the results returned from
-roccomp-, with the number of replications determined by the precision desired
for the power estimate.
Joseph Coveney
-------------------------------------------------------------------------------
Gregory Daniel wrote:
> Dear Statlist,
>
> I am using -roccomp- to compare areas under the ROC curve estimated from
> two independent samples. Can anyone point me in the right direction for
> appropriate power and sample size calculations for this test?
>
> Greg
>
*
* 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/
