Fig. 3From: Joint use of over- and under-sampling techniques and cross-validation for the development and assessment of prediction modelsCross-validated AUC for different sample sizes and classification rules obtained on simulated data. AUC obtained with different classification rules for simulated data with 10 variables (simulated independently from a Gaussian distribution with zero mean and unit variance) and 2 CV folds. There were n=100, 500, 1,000 and 100,00 samplesBack to article page