Non-parametric estimation of ROC curve

Receiver operating characteristic (ROC) curve is widely applied in measuring discriminatory ability of diagnostic or prognostic tests. This makes ROC analysis one of the most active re-search areas in medical statistics. Many parametric and semiparametric estimation methods have been proposed for estimating the ROC curve and its functionals. In this paper, we pro-pose a fully nonparametric Bayesian bootstrap (BB) estimation method for ROC curve and its functionals. The BB method gives a bandwidth free automatically smooth estimate. The area under the curve (AUC) is used to measure the accuracy of different diagnostic meth-ods. The accuracy of the estimate of the ROC curve in the simulation studies is examined by the integrated absolute error (IAE). In comparison with other existing curve estimation methods, BB method performs well in terms of accuracy, robustness and simplicity. We also propose a procedure based on the BB approach to test the binormality assumption