Number of permuted and unpermuted covariates, number of permuted covariates classified as important (false positives, FP), and number of unpermuted covariates classified as important (UI). FP and UI v
Results of a Friedman test to compare feature selection methods in terms of classification accuracy across different datasets and prediction methods (the best average ranks for each row are shown in b
AUC of signature size in {4, 11, 15, 22, 24, 26, 27, 28, 29, 32, 34, 35, 36}, combined with four classification algorithms in a 10-fold cross-validation. For each classification method, we highlighted