Quality Report

Clicking on the Quality Report icon shows performance indicators for each of the selected models.

Structure

Each model is evaluated for all available subsets (these are typically runs or experiments) within the training/validation and test set. Each subset (run/experiment) is displayed within a different tab.

Note

The tab name is identical to the subset name being summarized.

Each row represents a different model.

Errors and estimators types

The following errors and estimators are provided for each available model-subset pair for both the training/validation and the test set.

Table 10. Errors/Estimators types

Tag	Name
BIC	Bayesian information criterion
AICC	AIC with a correction for small sample sizes
AIC	Akaike information criterion
NRMSE	Normalized root mean square error
R2	Coefficient of determination (\(R^{2}\))
MallowsCp	Mallows’s \(C_{p}\)
FPE	Akaike’s Final Prediction Error for estimated model
MeanError	Mean error
StdError	Standard error
MaxAbsError	Maximum absolute error
MinAbsError	Minimum absolute error
SSE	Sum of squared estimate of errors
RMSE	Root mean square error
MSE	Mean squared error
NDIE	—
SAE	Sum absolute error
MAE	Mean absolute error
NMAE	Normalized mean absolute error
OUTPUTVARIANCE	Output variance
MODELOUTPUTVARIANCE	Model output variance

Note

If an error occurs during calculation due to zero division, etc. NaN (not a number) will be displayed in the performance table.