RFgroove (version 1.1)

selectGroup: Grouped variable selection procedure

Description

A grouped backward variable selection procedure.

Usage

selectGroup(design, ydata, varNames, nvarGroup,  typeRF = ifelse(is.factor(ydata), "classif", "reg"),  verbose = TRUE, ntree = 500, ...)

Arguments

design

The design matrix.

ydata

The outcome data. Must be a factor for classification.

varNames

The vector of the group names.

nvarGroup

The vector of the number of variables in each group.

typeRF

The type of forest we want to construct, ‘classif’ for classification or ‘reg’ for regression.

verbose

Should the details be printed.

ntree

The number of trees in the forests (default: 500).

...

optional parameters to be passed to the ‘varImpGroup’ function.

Value

An object of class fRFE which is a list with the following components:

nselected

The number of selected groups ;

selection

The selected groups ;

selectionIndexes

The indexes of selected groups in the input matrix ‘design’ ;

error

The prediction error computed in each iteration of the backward procedure ;

typeRF

The type of the forests, classification or regression ;

ranking

The final ranking of the groups ;

rankingIndexes

The final ranking indexes of the groups.

References

Gregorutti, B., Michel, B. and Saint Pierre, P. (2015). Grouped variable importance with random forests and application to multiple functional data analysis, Computational Statistics and Data Analysis 90, 15-35.

See Also

selectLevel,selectFunctional,varImpGroup

Examples

  cat("\n\nClassification\n")
  data(toyClassif)
  attach(toyClassif)
  
  cat("Case 1\n")
  nvarGroup <- c(2,1,6); names(nvarGroup) <- paste("G", 1:length(nvarGroup), sep="")
  summary(varSel <- selectGroup(design=X, ydata=Y, nvarGroup=nvarGroup, 
                                verbose=TRUE, ntree=500, normalize=FALSE))
  plot(varSel)

  cat("Case 2\n")
  nvarGroup <- rep(1,9); names(nvarGroup) <- paste("G", 1:length(nvarGroup), sep="")
  summary(varSel <- selectGroup(design=X, ydata=Y, nvarGroup=nvarGroup, 
                                verbose=TRUE, ntree=500, normalize=FALSE))
  plot(varSel)

  detach(toyClassif)