Anomano Discriminant Analysis

Using the example shown on startup: How well do MeanSpinkterPressure and LengthHPZ separate the groups of Indikation, for the pair defäk(ation) vs. inkon(tinenz)? The most sophisticated Random Forest classifier can have a lower ROC than the simpler methods, because it is less biased. The prediction uses a 10-fold cross-validation. The ROC curve is based on the predictions of the selected pair of groups, which are out-of-sample predictions, i.e. there is no self-referential prediction. The slider bars allow you to remove outliers and simplify the groups by removing rare indications.