ABSTRACT
Receiver operating characteristic (ROC) curve analysis is a popular tool for visualizing, organizing and selecting classi?ers based on their classification accuracy. Recently, the ROC methodology has been extensively adapted to medical areas heavily dependent on screening and diagnostic tests, in particular, radiology, bioinformatics, epidemiology, and laboratory testing. ROC curve analysis can also be applied generally for evaluating the accuracy of goodness-of-fit tests of statistical models (e.g., logistic regression) that classifies subjects into two categories such as diseased or non-diseased (e.g., in the context of a linear discriminant analysis). This chapter outlines the following ROC curve topics: ROC Curve Inference, Area under the ROC Curve, ROC Curve Analysis and Logistic Regression, Best Combinations Based on Values of Multiple Biomarkers. These themes are considered in the contexts of definitions, parametric/nonparametric estimation/testing and the relevant literature in the field of the ROC curve analyses.
