Title: An index to analyze a portion of the ROC curve
Authors: Maria del Carmen Pardo - Complutense University of Madrid (Spain)
Alba Franco-Pereira - Universidad Complutense de Madrid (Spain) [presenting]
Abstract: In clinical epidemiology, the partial Area Under the Curve (pAUC) at a high specificity threshold has been proposed as an indicator of test accuracy when the ROC curve is asymmetric. For example, acceptable specificities are high for early cancer detection tests. A lower specificity for a large population leads to many more falsely classified non-diseased subjects who may have to undergo a more invasive test subsequently. It is thus desired to compare screening markers at a higher range of specificities. The pAUC, which summarizes part of the ROC curve in the range of desired specificities, use to be the alternative to AUC. Several methods have been developed as alternative to pAUC but most of them assume the data have an underlying normal distribution. We propose a nonparametric partial summary index and its properties are explored by simulations. Finally, the new index is illustrated with a real data set.