Model Building - C&RT

On the Model building tab, select the C&RT check box and click the C&RT button to display this dialog box.

Stopping option for pruning. Specifies the stopping rule for the pruning computations. You can choose Misclassification error, Deviance, or FACT style direct stopping.

Minimum n per node. Specifies a minimum n per node when pruning should begin; this value controls when split selection stops and pruning begins.

Minimum child node size to stop. Use this option to control the smallest permissible number in a child node for a split to be applied. While the Minimum n per node parameter determines whether an additional split is considered at any particular node, the Minimum child node size to stop parameter determines whether a split will be applied, depending on whether any of the two resultant child nodes will be smaller (have fewer cases) than n as specified via this option.

Maximum number of nodes. Use this option to control the maximum number of nodes.

Maximum number of levels in tree. Use this option to control the maximum number of levels in the tree.

V-fold cross-validation. Performs V-fold cross-validation of the final tree (but not the full tree sequence; use the General C&RT options to compute v-fold crossvalidation estimates for the entire tree-sequence); in V-fold cross-validation, random samples are generated from the learning sample to provide an estimate of the CV cost for each classification tree in the tree sequence. Note that in data mining applications with large data sets, V-fold cross-validation may require significant computing time. Say True or False.

Prior class probabilities. Specifies how to determine the a-priori classification probabilities. You can set them equal or estimate them.

OK. Click this button to finalize your changes, close this dialog, and return to the Model building step.

Cancel. Click this button to cancel your changes and return to the Model building step.