IBM® Db2® for z/OS® models - Regression Tree

A regression tree is a tree-based algorithm that splits a sample of cases repeatedly to derive subsets of the same kind, based on values of a numeric target field. As with decision trees, regression trees decompose the data into subsets in which the leaves of the tree correspond to sufficiently small or sufficiently uniform subsets. Splits are selected to decrease the dispersion of target attribute values, so that they can be reasonably well predicted by their mean values at leaves.