Tree-AS node - growing

Use the growing options to fine-tune the tree-building process.

Record threshold for switching from p-values to effect sizes Specify the number of records at which the model will switch from using the P-values settings to the Effect size settings when building the tree. The default is 1,000,000.

Significance level for splitting Specify the significance level (alpha) for splitting nodes. The value must be between 0.01 and 0.99. Lower values tend to produce trees with fewer nodes.

Significance level for merging Specify the significance level (alpha) for merging categories. The value must be between 0.01 and 0.99. This option is not available for Exhaustive CHAID.

Adjust significance values using Bonferroni method Adjust significance values when you are testing the various category combinations of a predictor. Values are adjusted based on the number of tests, which directly relates to the number of categories and measurement level of a predictor. This is generally desirable because it better controls the false-positive error rate. Disabling this option increases the power of your analysis to find true differences, but at the cost of an increased false-positive rate. In particular, disabling this option may be recommended for small samples.

Effect size threshold (continuous targets only) Set the effect size threshold to be used when splitting nodes and merging categories; when using a continuous target. The value must be between 0.01 and 0.99.

Effect size threshold (categorical targets only) Set the effect size threshold to be used when splitting nodes and merging categories; when using a categorical target. The value must be between 0.01 and 0.99.

Allow resplitting of merged categories within a node The CHAID algorithm attempts to merge categories in order to produce the simplest tree that describes the model. If selected, this option enables merged categories to be resplit if that results in a better solution.

Significance level for grouping leaf nodes Specify the significance level that determines how groups of leaf nodes are formed or how unusual leaf nodes are identified.

Chi-square for categorical targets For categorical targets, you can specify the method used to calculate the chi-square statistic.

• Pearson This method provides faster calculations but should be used with caution on small samples.
• Likelihood ratio This method is more robust than Pearson but takes longer to calculate. For small samples, this is the preferred method. For continuous targets, this method is always used.