The model can be displayed either as a tree, as in Figure 1 (left), or as a rectangular partition of the feature space, as in Figure 1 (right): The first split in the variable friends_smoke partitions the entire sample, while the second split in the variable alcohol_per_month further partitions only those subjects whose value for the variable friends_smoke is ”one or more”. The partition representation in Figure 1 (right) is even better suited than the tree representation to illustrate that recursive partitioning creates nested rectangular prediction areas corresponding to the terminal nodes of the classification tree. Details about the prediction rules derived from the partition are given below.
