PROC HPFOREST is trained with DEPTH=10 on a dataset with 100 predictor variables. A colleague suggests reducing DEPTH=4. What is the primary effect of reducing tree depth in a random forest?
Select an answer to reveal the explanation.
Short Explanation and Infographic
Here's the deal — b is correct because tree depth controls the bias-variance tradeoff in random forest. Very deep trees (high DEPTH=) have low bias per tree but high variance (overfit to bootstrap sample noise).
Full explanation below image
Full Explanation
B is correct because tree depth controls the bias-variance tradeoff in random forest. Very deep trees (high DEPTH=) have low bias per tree but high variance (overfit to bootstrap sample noise). Reducing depth increases bias (each tree captures less complex patterns) but reduces variance. The ensemble average then reflects the bias-variance tradeoff of the base trees. A is wrong; training time decreases with lower depth. C is wrong; depth affects the ensemble's overall bias-variance balance. D is wrong; VARS_TO_TRY= controls variables per split, independent of DEPTH=.