Mean absolute scaled error

Mean absolute scaled error is a number that measures how well a model or prediction did. Generally, the lower the number, the better the prediction was. It takes into account how far away the prediction was from the actual number (error) and also how it compares to other predictions that have been made. For example, if the actual value was 100 and the model's prediction was 60, then the error would be 40. However, if the second prediction was also 40, then the mean absolute scaled error would be 0 because the both predictions are equal distances away from the actual value.