Decision Tree Properties

The following properties are available for the decision tree:
Name
enables you to specify the name for this model.
Maximum branches
specifies the maximum number of branches allowed when splitting a node.
Maximum levels
specifies the maximum depth of the decision tree.
Leaf size
specifies the minimum number of observations allowed in a leaf node.
Response bins
specifies the number of bins used to categorize a measure response variable.
Predictor bins
specifies the number of bins used to categorize a predictor that is a measure variable.
Pruning
specifies the aggressiveness of the tree pruning algorithm. A more aggressive algorithm creates a smaller decision tree. Larger values are more aggressive.
Rapid growth
enables you to use the information gain ratio and k-means fast search methods for decision tree growth. When disabled, the information gain and greedy search methods are used, which generally produce a larger tree and require more time to create.
Include missing
enables you to include observations with missing values. For category variables, a missing value is assigned to its own level. For measure variables, a missing value is assigned to the smallest available machine value (negative infinity).
Reuse predictors
allows more than one split in the same branch based on a predictor.
Frequency
specifies whether nodes report how many observations they contain or what percentage of the observations they contain.
Assessment
  • Use default number of bins specifies whether you want to use the default number of bins or to set your own value. By default, measure variables are grouped into 20 bins.
  • Number specifies the number of bins to use when the Use default number of bins property is not selected. You must specify an integer value between 5 and 100.
  • Prediction cutoff specifies the value at which a computed probability is considered an event.
  • Tolerance specifies the tolerance value that is used to determine the convergence of the iterative algorithm that estimates the percentiles. Specify a smaller value to increase the algorithmic precision.
Show diagnostic plots
specifies whether the Leaf Statistics and Assessment windows appear in the model pane.
Show tree overview
displays the tree overview. The tree overview enables quick navigation of large decision trees. When you zoom in to view a specific area of the decision tree, the tree overview shows the entire decision tree and highlights the area that you are viewing. You can click and drag the highlighted area to change the display of the decision tree. Click the zoom to fit icon icon in the upper left corner of the tree overview to view the entire decision tree. Click the minimize tree icon icon in the upper left corner of the tree overview to minimize the tree overview.
Copyright © SAS Institute Inc. All rights reserved.