The percentage representation of the estimated model is calculated
as the normalized mean error for the prediction (FIT1).
The loss function or error function (V) is the numeric value of the mean squared error (MSE) that is computed with
the estimation data set.
Finally, the generalization error value is the numeric value of the normalized sum of squared errors (NSSE) that is computed with the validation data set
(NSSE1) and with the test data set (NSSE2).