A data scientist uses PROC HPLOGISTIC and requests the ROC curve output. Which ODS output object contains the ROC curve data points (sensitivity and 1-specificity pairs)?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — you need the ODS OUTPUT statement with the ROCCurve destination because that captures the ROC curve data table when PROC HPLOGISTIC requests ROC statistics. That table gives you the sensitivity and one-minus-specificity pairs at each threshold.
Full Explanation
The ODS OUTPUT statement with the ROCCurve destination is correct because it captures the ROC curve data table produced by PROC HPLOGISTIC when ROC statistics are requested. This table contains the sensitivity and one-minus-specificity pairs at each predicted probability threshold. The OUTROC option in the MODEL statement is wrong because it is valid syntax in PROC LOGISTIC but not PROC HPLOGISTIC. The ODS OUTPUT statement with the Association destination is wrong because it captures association statistics such as the c-statistic but not the full ROC curve point pairs. The claim that PROC HPLOGISTIC does not produce ROC curves and that you must use PROC LOGISTIC instead is wrong because PROC HPLOGISTIC does produce ROC curve information.