95% confidence interval for lroc?

[Robin Fatch]

Hi,

I am using lroc after different logistic regression models to estimate the area under the ROC curve. I was wondering if there is a way to include/calculate a 95% confidence interval for the AUC...?

Thanks,
Robin

logistic y c.var1 i.var2 i.var3, base

lroc, nograph 

[Joseph Coveney]

I am using lroc after different logistic regression models to estimate the area under the ROC curve. I was wondering if there is a way to include/calculate a 95% confidence interval for the AUC...?

Yes: substitute the following two lines of code (in red) for your lroc , nograph line.

logistic y c.var1 i.var2 i.var3, base
predict double xb, xb
roctab foreign xb

[Joseph Coveney]

Sorry, those lines were copied from an illustration using the auto dataset.

Use the following final line for your displayed code.

roctab y xb

Here's the illustration, if you're curious.

.ÿ
.ÿversionÿ16.1

.ÿ
.ÿclearÿ*

.ÿ
.ÿsetÿseedÿ`=strreverse("1560592")'

.ÿ
.ÿquietlyÿsysuseÿauto

.ÿ
.ÿsummarizeÿrep78,ÿmeanonly

.ÿquietlyÿreplaceÿrep78ÿ=ÿruniformint(r(min),ÿr(max))ÿifÿmi(rep78)

.ÿ
.ÿquietlyÿlogisticÿforeignÿc.gear_ratioÿi.rep78,ÿbase

.ÿlrocÿ,ÿnograph

Logisticÿmodelÿforÿforeign

numberÿofÿobservationsÿ=ÿÿÿÿÿÿÿ71
areaÿunderÿROCÿcurveÿÿÿ=ÿÿÿ0.9652

.ÿ
.ÿtempnameÿarea

.ÿscalarÿdefineÿ`area'ÿ=ÿr(area)

.ÿ
.ÿ*
.ÿ*ÿBeginÿhere
.ÿ*
.ÿquietlyÿpredictÿdoubleÿxb,ÿxb

.ÿroctabÿforeignÿxb

ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿROCÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ-AsymptoticÿNormal--
ÿÿÿÿÿÿÿÿÿÿÿObsÿÿÿÿÿÿÿAreaÿÿÿÿÿStd.ÿErr.ÿÿÿÿÿÿ[95%ÿConf.ÿInterval]
ÿÿÿÿÿ------------------------------------------------------------
ÿÿÿÿÿÿÿÿÿÿÿÿ71ÿÿÿÿÿ0.9652ÿÿÿÿÿÿÿ0.0217ÿÿÿÿÿÿÿÿ0.92272ÿÿÿÿÿ1.00000

.ÿ
.ÿassertÿr(area)ÿ==ÿ`area'

.ÿ
.ÿexit

endÿofÿdo-file


.