The performing of several diagnostic tests on an individual provide multiple measures that can be combined in order to optimize diagnosis accuracy.
Di erent approaches for choosing linear combinations of diagnostic tests that maximize the Area Under Receiver Operating Characteristic Curve (AUC) have been proposed in the literature.
A step by step algorithm for estimating the parameters of models is introduced. Moreover, extensive simulations in di erent settings have been considered in order to study the performance of the algorithm. The proposed approach is applied to a prostate cancer dataset.
© 2008-2024 Fundación Dialnet · Todos los derechos reservados