Fitness Functions for Time Series Prediction

Choosing the Fitness Function

For Time Series Prediction problems, the fitness functions of GeneXproTools 4.0 are organized as follows:

Fitness functions based on the absolute error:

Absolute Error with Selection Range
Absolute/Hits
MSE (mean squared error)
RMSE (root mean squared error)
MAE (mean absolute error)
RSE (relative squared error)
RRSE (root relative squared error)
RAE (relative absolute error)

Fitness functions based on the relative error:

Relative Error with Selection Range
Relative/Hits
rMSE (relative MSE)
rRMSE (relative RMSE)
rMAE (relative MAE)
rRSE (relative RSE)
rRRSE (relative RRSE)
rRAE (relative RAE)

Fitness functions based on correlation coefficients:

R-square
Correlation Coefficient

Fitness functions with parsimony pressure:

And all the above with parsimony pressure

And user defined fitness functions:

Custom Fitness Function

Home | Contents | Previous | Next