What does MAD measure?

What does MAD measure?

• A. The size of the forecast error in units
• B. The size of the forecast error in percentage terms
• C. The average forecast error
• D. The maximum forecast error

Answer: (A)

MAD captures the typical size of forecast errors, expressed in the same units as the data. It is calculated as the average of the absolute differences between what was forecast and what actually happened, so it reflects how large errors tend to be without regard to whether they’re over- or under-predictions. This makes MAD a unit-consistent measure of forecast accuracy, telling you, on average, how far off forecasts are.

Among the options, describing MAD as the size of the forecast error in units fits best because it emphasizes the magnitude of errors in the data’s own units. The other choices describe different ideas: using percentage terms would be a percentage-based error measure (like MAP), the average error could imply a signed average (which MAD avoids by using absolute values), and the maximum error refers to the worst single error rather than the typical size of errors on average.