Classification Accuracy and Cut-Point Selection

• Activity Number: 127
• Type: Contributed
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Consulting
• Author(s): Xinhua Liu*+
• Companies: Columbia University
• Keywords: Classification accuracy ; Concordance probability ; Cut point ; Sensitivity ; Specificity

Abstract

In biomedical research and practice, quantitative tests or biomarkers are often used for diagnostic or screening purposes, with a cut point established on the quantitative measurement to aid binary classification. We introduce an alternative to the traditional methods based on the Youden index and the closest-to-(0, 1) criterion for threshold selection. A concordance probability evaluating the classification accuracy of a dichotomized measure is defined as an objective function of the possible cut point. A non-parametric approach is used to search for the optimal cut point maximizing the objective function. The procedure is shown to perform well in a simulation study. Using data from a real-world study of arsenic-induced skin lesions, we apply the method to a measure of blood arsenic levels, selecting a cut point to be used as a warning threshold.