Why might one prefer to use MAE over MSE in evaluating a regression model?

  • MAE considers the direction of errors
  • MAE gives more weight to larger errors
  • MAE is less sensitive to outliers
  • MAE is more computationally expensive
One might prefer to use Mean Absolute Error (MAE) over Mean Squared Error (MSE) because MAE is less sensitive to outliers. While MSE squares the differences and thus gives more weight to larger errors, MAE takes the absolute value of the differences, providing an equal weighting. This makes MAE more robust when there are outliers or when one doesn't want to overly penalize larger deviations from the true values.
Add your answer
Loading...

Leave a comment

Your email address will not be published. Required fields are marked *