Skip to content

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

A risk team chooses the penalty parameter lambda for an elastic net model used to predict loan defaults. Which procedure is most appropriate?

Lambda should be chosen by cross-validation, picking the value with the best validation performance. In-sample error always favors a near-zero penalty and overfitting, while cross-validation estimates out-of-sample error and balances bias and variance in the elastic net.

  1. AChoose lambda that minimizes in-sample error on the full training set
  2. BChoose lambda by cross-validation, selecting the value with the best validation performanceCorrect
  3. CSet lambda to zero to maximize flexibility
  4. DChoose the largest lambda so all coefficients equal zero

Explanation

Lambda is a hyperparameter not estimated by the penalized fit itself. Minimizing in-sample error would always choose lambda near zero and overfit. Cross-validation estimates out-of-sample performance for each lambda, so the best validation value is chosen.

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