Receiver Operating Characteristic ( ROC ) Curves
James A. Hanley
Abstract
James A. Hanley
Abstract
Abstract The receiver operating characteristic(ROC) curve describes the relationship between the false positive fraction and true negative fraction associated with a diagnostic test as the test threshold defining the boundary between individuals classified as cases or noncases is varied. The ROC curve avoids certain arbitrariness involved in the choice of a single overall index of accuracy. This entry describes how the ROC curve is derived from data, and reviews summary indices that may represent its performance. The comparison of ROC curves for competing tests is discussed, and appropriate software is identified.
OpenAlex reports 42 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Abstract The receiver operating characteristic(ROC) curve describes the relationship between the false positive fraction and true negative fraction associated with a diagnostic test as the test threshold defining the boundary between individuals classified as cases or noncases is varied. The ROC curve avoids certain arbitrariness involved in the choice of a single overall index of accuracy. This entry describes how the ROC curve is derived from data, and reviews summary indices that may represent its performance. The comparison of ROC curves for competing tests is discussed, and appropriate software is identified.
Key concepts: Receiver operating characteristic, Fraction (chemistry), Boundary (topology), Mathematics, Statistics, Arbitrariness, Mathematical analysis, Chromatography