Confidence Interval Estimation of the Area under the Receiver Operating Characteristic Curve in the Presence of Measurement Error
Date of Award
Master of Science
Epidemiology and Biostatistics
Dr. GuangYong Zou
Diagnosis of diseases is often based on biomarkers with continuous measurements. The discriminative ability of a biomarker can be depicted by a receiver operating characteristic (ROC) curve, which shows simultaneously the proportions of both ab normal and normal subjects correctly diagnosed at various cutoff points in the marker values. The area (A) under the ROC curve is commonly used to measure the ability of the marker to distinguish between two populations. Many biomarkers are subject to measurement error, which must be taken into account in statistical inference for A to avoid misleading results. Assuming a normal distribution for biomarker values, this thesis developed a confidence interval procedure for A adjusted for random mea surement error that can be quantified by an external reliability study. The basis of the new procedure is the method of variance estimates recovery. Simulation results show that this procedure outperformed the one based on the Delta method. The methodology is illustrated by a data set from a study using thiobarbituric acid reac tion substance to diagnose cardiovascular disease.
Li, Yanhong, "Confidence Interval Estimation of the Area under the Receiver Operating Characteristic Curve in the Presence of Measurement Error" (2009). Digitized Theses. 3846.