Figure 4

An example of a classification tree of outcomes y > 1 (class 1) and y < 1 (class 0) using all the parameters in the model except k . The items displayed in the nodes in the tree diagram are: the criterion for making the decision (e.g.: α _fn ≥ 0.1626), the predicted class for that terminal node (0 or 1) and the number of observations correctly classified to the class versus the number misclassified, achieved by cross-validation (e.g.: 2162/84). To proceed through the diagram at a given node move to the left branch if the stated condition is true (yes) and to the right if false (no).