522 – Bayesian Statistics with Biomedical Applications
Weight Optimization for Comparing Areas Under ROC Curve for a Repeated Marker Between Correlated Groups
Ping Xu
University of South Florida
Yougui Wu
University of South Florida
The area under the Receiver Operating Characteristic curve (AUC) is often used to evaluate the prognostic performance of a continuous biomarker. A non-parametric AUC estimate with optimal weights has been introduced in the literature to compare two biomarkers with repeated measurements. We modify this estimate to examine the AUC difference between two matched groups, taking into account not only the within-group correlation but also the between-group correlation. It is demonstrated how the Lagrange multiplier can be used as a strategy for finding the weights which minimize the variance function subject to constraints. We show substantial gains of efficiency by using our proposed weighting scheme when both the within-group and between-group correlations are high, and/or the disease incidence is small, which is the case for many longitudinal matched design studies. An illustrative example is presented to apply the proposed methodology to a thyroid function dataset. Simulation results suggest that the optimal weight performs well with a sample size as small as 50 per group.