Glossaria.net

Glossary Time-Series / Term

Mean Absolute Percentage Error

(MAPE) A measure of forecast accuracy calculated by taking the absolute value of the individual forecast errors of the time series, dividing by the actual value of the time series, multiplying by 100, and calculating the average of the resulting values. The MAPE is not related to the scale of the observations, so it is useful in comparing the errors between distinct variables that are not measured in the same units.

Permanent link Mean Absolute Percentage Error - Modification date 2019-12-22 - Creation date 2019-12-22


< Mean Absolute Error Glossary / Time-Series Mean Error >