For a binary classification problem, which splitting criterion does PROC HPFOREST use by default?
Select an answer to reveal the explanation.
Short Explanation
PROC HPFOREST defaults to Gini impurity reduction for classification. You pick the split that most reduces the chance of misclassifying a randomly chosen case.
Full Explanation
PROC HPFOREST defaults to Gini impurity reduction for binary classification. Gini measures the probability of misclassifying a randomly chosen element, and splits maximize Gini impurity reduction. Information gain or entropy reduction is used by ID3/C4.5. The chi-square test of independence criterion is used by CHAID. Reduction in mean squared error is for regression trees.