What does mean absolute error (MAE) measure?

Mean Absolute Error (MAE) is a statistical measure used to assess the accuracy of a forecasting method. It calculates the average of the absolute differences between the predicted values and the actual observed values. By focusing on the absolute values of errors, MAE avoids the problem of positive and negative errors cancelling each other out, allowing for a more straightforward interpretation of forecast accuracy.

In this context, the emphasis on 'average absolute' in the correct answer refers specifically to the averaging process of these absolute forecast errors. This makes MAE particularly useful because it provides a clear, parsimonious measure of forecasting performance in units that are directly interpretable, aligning the measure with the scale of the data being forecasted.

Other options include metrics that either deal with errors in percentage form, total errors without averaging, or emphasize only the maximum error, which fail to capture the comprehensive nature of the forecasting performance indicated by MAE. Therefore, the correct choice succinctly encapsulates the primary function of MAE in evaluating forecast accuracy.