Classification and Regression Trees

Free SOA Exam SRM (Statistics for Risk Modeling) lesson in Decision Trees. 20 min read, ~2,946 words.

Trees partition predictor space into rectangles, predicting one constant per region: the leaf mean of Y (regression) or the majority class (classification). Regression splits minimize residual sum of squares (RSS); classification splits minimize Gini index or cross-entropy. Classification error rate is too coarse for growing. Trees grow greedy and top-down...