The PARTITION statement specifies how observations in the input data set are to be logically partitioned into disjoint subsets for model training, validation, and testing. Models are fit and selected based on the training data. After you fit a model, you can use the validation and test sets to assess how the selected model generalizes on data that played no role in selecting the model.
The PARTITION statement divides the observations in the input data set into disjoint subsets for model training, validation, and testing. Various fit statistics are displayed in the new "Partition Fit Statistics" table.
The new CUTPOINT= option in the MODEL statement enables you to control the classification of events and nonevents.
The new CHOOSE=VALIDATE and STOP=VALIDATE options in the SELECTION statement use the validation data set during the selection process.
The AIC, BIC, and AICC criteria are added to the SELECT= option in the SELECTION statement.
The INEST= option in the PROC statement enables you to input your own starting values for the optimization. The OUTEST option adds a column that contains the parameter names to the "Parameter Estimates" ODS OUTPUT data.
The CTABLE option in the MODEL statement creates data for receiver operating characteristic (ROC) curves. The PRIOR= option in the MODEL statement specifies population prevalences that are used to adjust statistics displayed by the CTABLE option and by the PARTITION statement.
The POST keyword in the OUTPUT statement outputs the posterior probabilities that are specified in the PRIOR= option, and the ROLE keyword outputs the partition to which the observation is assigned.