GLOSSARY

Page tree

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 7 Next »

@wikipedia


The basic set of Model Validation metrics against certain dataset (could be the Training dataset or Validation dataset).

There is no complete set of numbers to define the  Goodness of fit and different studies may use different sets of Goodness of fit metrics but the basic set is provided by the following table:


Mean Square Deviation (MSD)MSD
Root Mean Square Deviation (RMSD)RMSD
Average Relative Error (ARE)ARE
Average Absolute Relative Error (AARE)AARE
Maximum Relative Error (MAXRE)MAXRE
Maximum Absolute Relative Error (MAXARE) MAXARE
 Coefficient of determination (R2)R2
Pearson correlation coefficient (ρP) ρP
Correlation skewnessS3
Validation Cross-Plot


The most widely used set in engineering practice is given by:

Root Mean Square Deviation (RMSD)RMSD
Average Absolute Relative Error (AARE)AARE
Maximum Absolute Relative Error (MAXARE) MAXARE
 Coefficient of determination (R2)R2
Pearson correlation coefficient (ρP) ρP
Correlation skewnessS3

Validation Cross-Plot



See also

Formal science / Mathematics / Statistics / Statistical Metric 

[ Mean Square Deviation (MSD) = Mean Square Error (MSE) ]

[ Root Mean Square Deviation (RMSD) = Root Mean Square Error (RMSE) ]

[ Average Relative Error (ARE) = Average Percentage Error (APE) ]

[ Average Absolute Relative Error (AARE) = Average Absolute Percentage Error (AAPE)]

[ Maximum Relative Error (MAXRE) = Maximum Percentage Error (MAXPE) ]

[ Maximum Absolute Relative Error (MAXARE) = Maximum Absolute Percentage Error (MAXAPE) ]

[ Coefficient of determination (R2) ][ Pearson correlation coefficient (ρP) ][ Correlation skewness ]

  • No labels