TAAFT
Free mode
100% free
Freemium
Free Trial
Create tool

Mean Absolute Error

[min ˈæbsəlut ˈɛrər]
Machine Learning
Last updated: December 9, 2024

Definition

The average of the absolute differences between predicted and actual values. Treats all errors equally regardless of direction.

Detailed Explanation

MAE calculates the average magnitude of errors without considering their direction, using absolute values instead of squares. This makes it less sensitive to outliers than MSE/RMSE and provides a linear loss function. It's particularly useful when outliers should not have extra influence.

Use Cases

Applicable in inventory forecasting, budget planning, and resource allocation where the magnitude of error is more important than its direction.

Related Terms