Home / ML Associate practice test / Model Development / Question

Databricks Machine Learning Associate · free practice question

A forecasting team wants an evaluation metric that penalizes large prediction errors more strongly than small ones because a few bad misses are very costly…

Exam-style question on Model Development from the free ML Associate practice test, with the answer and a full explanation below.

Updated November 10, 2026

Model Development

A forecasting team wants an evaluation metric that penalizes large prediction errors more strongly than small ones because a few bad misses are very costly. Which regression metric aligns best with that preference?

  1. AMean absolute error, because it weights all residual magnitudes linearly and does not emphasize very large misses as strongly
  2. BMedian absolute error, because it intentionally downweights large errors by focusing on the middle residual
  3. CRoot mean squared error, because squaring residuals increases the penalty on larger prediction errors
  4. DR-squared, because it reports explained variance rather than expressing error in the target units directly
  5. EExplained variance score, because it compares target dispersion but is not primarily an error-magnitude measure
Answer & explanation

Correct answer: C

WHY C: RMSE squares residuals before averaging, so larger errors receive greater emphasis than smaller ones. WHY NOT A: MAE is linear and less sensitive to large misses. WHY NOT B: Median absolute error is even more robust to outliers. WHY NOT D: R-squared is not a direct error magnitude in original units. WHY NOT E: Explained variance is not the best match for this cost preference.

Take the full ML Associate practice test free →

Would you pass today?

This question is one of 45 in the free, timed Databricks Machine Learning Associate practice test. Take it and get a section-by-section score.

Start the free test →