Recursive Feature Elimination for SCE models to identify the most important predictors.
RFE_SCE(Training_data, Testing_data, Predictors, Predictant, Nmin, Ntree,
alpha = 0.05, resolution = 1000, step = 1, verbose = TRUE,
parallel = TRUE)RFE results with performance metrics and importance scores.
Training dataset
Testing dataset
Character vector of predictor names
Character vector of predictant names
Minimum samples per node
Number of trees
Significance level (default: 0.05)
Resolution for splitting (default: 1000)
Number of predictors to remove per iteration (default: 1)
Print progress (default: TRUE)
Use parallel processing (default: TRUE)
Plot_RFE, SCE, importance