Skip to content

FRM Part I · FRM Exam Part I · Machine-Learning Methods

A random forest differs from simply bagging many fully grown decision trees because, at each split, a random forest:

A random forest considers only a random subset of predictors at each split. This decorrelates the bootstrapped trees, so averaging their predictions reduces variance more than plain bagging. Sequential residual fitting is boosting, not a random forest.

  1. AConsiders only a random subset of the predictors as split candidates, reducing correlation among the treesCorrect
  2. BFits each tree sequentially to the residuals of the previous tree
  3. CUses all predictors but weights misclassified observations more heavily
  4. DReplaces the trees' majority vote with a single linear regression

Explanation

Random forests add random predictor selection at each split on top of bootstrap sampling. This decorrelates the trees, so averaging reduces variance more effectively. Sequential residual fitting and reweighting describe boosting, not random forests.

Did you get it right without looking?

One question tells you little. A timed set on Machine-Learning Methods shows your real accuracy, how long you take and where you lose marks.

Practise Machine-Learning MethodsFree diagnostic mock

More Machine-Learning Methods questions