Figure 3From: A Wild Bootstrap approach for the selection of biomarkers in early diagnostic trialsEmpirical type-I error for varying AUC’s. Empirical type-I error of the different approaches for the standard scenario (see text) with varying AUC’s.Back to article page