Mean relative error
AmP estimation  


For each entry a Mean Relative Error (MRE) is computed to help judge the goodness of fit.
Definition
To compute the MRE we start by computing the Relative Error (RE) for each data set. The RE of a data set is computed in this way
where i refers to the data set and j to a given point in data set i. d_{ij} stands for the data, p_{ij} for the model prediction and w_{ij} for the associated weight coefficient. Finally, d_{i} in the denominator represents the average of all data points in set i (d_{ij}):
In the case the sum of the weights w_{i} is equal to zero we simply take the relative error to be zero. The implication is that this data set does not contribute to the MRE.
MRE is then the mean of all RE_{i} with w_{i} greater than 0.
where n' is the number of data sets with w_{i} greater than 0.
The zerovariate case
In the case of a zerovariate data set the relative error simplifies to
and the connection with the formal definition of relative is more easily seen.