General CHAID Models Quick Specs - Validation Tab

Select the Validation tab of the General CHAID Models Quick specs dialog box to access two types of cross-validation methods: V-fold cross-validation and Test sample.

Element Name Description
V-fold cross-validation Select the V-fold cross-validation check box to use this method. V-fold cross-validation is particularly useful when no test sample is available and the learning sample is too small to have the test sample taken from it.
Seed for random number generator The positive integer value entered here is used as the seed for a random number generator that produces v-fold random subsamples from the learning sample to test the predictive accuracy of the computed classification trees.
V-fold cross-validation; v-value The value entered here determines the number of cross-validation samples that will be generated from the training data to provide an estimate of the cross-validation error rate for the CHAID tree model that is constructed. See also the Introductory Overview for details.

See also Avoiding over-fitting: Pruning, cross-validation, and V-fold cross-validation in the GC&RT Introductory Overview for details.

Test sample Click this button to display the Cross-Validation dialog box. Here, you can specify a subsample of cases for estimating the accuracy of the classifier or prediction by switching on or off the Test sample option as well as selecting a variable that will be used as the sample identifier variable.

If a Test sample is specified, various options in the Results dialog box will compute separate statistics (predicted values or classifications, risk estimates, etc.) for the learning and the test sample.