Skip to content

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

A K-nearest neighbors classifier predicts whether a firm will default using two features: leverage (ratio, range 0 to 5) and market capitalization in USD millions (range 100 to 50,000), both unscaled. Which statement is most accurate?

Market capitalization will dominate the distance calculation because its numeric range is vastly larger than leverage, so features should be standardized before using KNN. KNN uses raw distances and learns no feature weights, so unscaled inputs distort which observations count as nearest neighbors.

  1. ADistances will be dominated by market capitalization, so features should be standardized before applying KNNCorrect
  2. BDistances will be dominated by leverage because it is a smaller number
  3. CScaling is unnecessary because KNN learns feature weights during training
  4. DScaling matters only for the choice of K, not for neighbor identification

Explanation

Euclidean distance uses raw units, so the feature with the largest numeric range, market cap, dominates neighbor selection. Standardizing puts features on comparable scales. KNN has no training stage that learns weights.

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