Question

The most commonly used criteria for measuring forecast error is

Answer 1

The most common method used for measuring forecast error is - Root Mean Square error.

Mean Squared Error is synonymously used with Mean Absolute Deviation.

The problem with the MSE is that the square puts a very high weight on large deviations, so the MSE-optimal forecast will have fewer large errors but may have much more small errors than a MAD-optimal forecast.

However, if there is only one-time series, it is better to use a mean absolute error (MAE). MAE is attractive as it is simple to understand and calculate.