The model comparison node rates the model performance based on its accuracy, misclassification, profit or loss, and also by the Kolmogorov-Smirnov (KS) statistic. Nevertheless, the accuracy of the model refers to the correct prediction decisions, while the misclassification refers to incorrect prediction decisions.
The choice of fit statistics has to depend on the predictions of interest. Referring to the table below for the fit statistic of each model:
The choice of fit statistics has to depend on the predictions of interest. Referring to the table below for the fit statistic of each model: