The LOGSELECT Procedure

Specifying a PARTITION statement modifies the display of many tables by adding separate rows or columns for the training, validation, and test data tables, and adds statistics to the "Fit Statistics" table. These additional statistics are useful for assessing the model and should be very similar for the different roles when the training data are representative of the other data partitions: average square error, misclassification rate, and the difference of means. The , maximum-rescaled , and McFadden’s are also computed. For more information, see the section Model Fit and Assessment Statistics.