Area under the Receiver Operating Characteristic Curve (AUC-ROC)

This chapter 17 focuses on investigating the area under the receiver operating characteristic curve (AUC-ROC). In addition to the Summary, there are four sections. Section 17.1 introduces the AUC-ROC, including the AUC-ROC graph, score, and the diagnostic test using AUC-ROC. Section 17.2 introduces three approaches for calculating the AUC-ROC: (1) using an iterative maximum likelihood estimation, (2) using the Wilcoxon test statistic, and (3) simply using trapezoid integration. Section 17.3 demonstrates how to obtain the AUC-ROC values via the SIAMCAT and mltools packages, respectively. Section 17.4 provides comprehensive remarks on AUC-ROC properties, advantages, and disadvantages as a performance metric. Particularly, this section discusses why we consider the AUC-ROC properties and limitations when applied to microbiome research.