Which metric is defined by the average of squared errors, punishing larger errors more than smaller ones?

Which metric is defined by the average of squared errors, punishing larger errors more than smaller ones?

• A. MAPE
• B. RMSE
• C. MAE
• D. MSE

Answer: (D)

The key idea here is a loss measure that grows faster as errors get larger. When you square each error, big mistakes become much bigger even than small ones, and averaging those squared errors across all predictions gives the mean squared error. This makes the metric very sensitive to large deviations, which is desirable when you want to strongly discourage large mistakes. This description points to mean squared error, as it is exactly the average of squared differences between predicted and true values. It’s different from mean absolute error, which averages absolute (not squared) errors; from mean absolute percentage error, which uses relative percentage errors; and from root mean squared error, which is the square root of the average of squared errors (retaining the same idea but changing the scale).

The key idea here is a loss measure that grows faster as errors get larger. When you square each error, big mistakes become much bigger even than small ones, and averaging those squared errors across all predictions gives the mean squared error. This makes the metric very sensitive to large deviations, which is desirable when you want to strongly discourage large mistakes.

This description points to mean squared error, as it is exactly the average of squared differences between predicted and true values. It’s different from mean absolute error, which averages absolute (not squared) errors; from mean absolute percentage error, which uses relative percentage errors; and from root mean squared error, which is the square root of the average of squared errors (retaining the same idea but changing the scale).