Absolute error is defined as the absolute value of the difference between the forecast value and actual value. Let's imagine a scenario as follows:
|
|
Actual value |
Predicted value |
Error |
Absolute error |
|
Data point 1 |
100 |
120 |
20 |
20 |
|
Data point 2 |
100 |
80 |
-20 |
20 |
|
Overall |
200 |
200 |
0 |
40 |
In the preceding scenario, we see that the overall error is 0 (as one error is +20 and the other is -20). If we assume that the overall error of the model is 0, we are missing out the fact that the model is not working well on individual data points.
Hence, in order to avoid the issue of a positive error and negative error canceling each other out and thus resulting in minimal error, we consider the absolute ...
