DTREG analyzes data files and generates classification and regression decision trees to model the data. V-fold cross validation and pruning are used to generate the optimal size tree for best generalization to other data. Once a tree has been created, it can be used to predict the values of other cases.