R Graphical Manual

Browse All

Last data update: 2014.03.03

R: Grouped variable selection procedure

selectGroup

R Documentation

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.

Author(s)

Baptiste Gregorutti

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.

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)

Results


R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> library(RFgroove)
Loading required package: randomForest
randomForest 4.6-12
Type rfNews() to see new features/changes/bug fixes.
Loading required package: wmtsa
Loading required package: fda
Loading required package: splines
Loading required package: Matrix

Attaching package: 'fda'

The following object is masked from 'package:graphics':

    matplot

> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/RFgroove/selectGroup.Rd_%03d_medium.png", width=480, height=480)
> ### Name: selectGroup
> ### Title: Grouped variable selection procedure
> ### Aliases: selectGroup
> 
> ### ** Examples
> 
>   cat("\n\nClassification\n")


Classification
>   data(toyClassif)
>   attach(toyClassif)
>   
>   cat("Case 1\n")
Case 1
>   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))
normalize = TRUE 
G1 G2 G3 
 2  1  6 
Classification backward selection.
Splitting data into a training and a testing set...
Survival indexes : 1 2 3 4 5 6 7 8 9 
   G1    G2    G3 
0.167 0.106 0.001 
G3 eliminated. 2 remaining groups of variables.	 Error = 0.12 


Survival indexes : 1 2 3 
   G1    G2 
0.188 0.131 
G2 eliminated. 1 remaining groups of variables.	 Error = 0.13 


Survival indexes : 1 2 
G1 eliminated. No remaining groups of variables.	 Error = 0.37 


Ending...
 3 selected variables:
 G1 G2 G3 



 --- 				 ---
 --- 	Summary functional RFE	 ---
 --- 				 ---


Number of selected variables using a validation set: 3 

Selected variables:
G1 G2 G3 

Validation error for the best model: 0.1226 

>   plot(varSel)
> 
>   cat("Case 2\n")
Case 2
>   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))
normalize = FALSE 
G1 G2 G3 G4 G5 G6 G7 G8 G9 
 1  1  1  1  1  1  1  1  1 
Classification backward selection.
Splitting data into a training and a testing set...
Survival indexes : 1 2 3 4 5 6 7 8 9 
    G3     G2     G1     G6     G7     G5     G9     G4     G8 
 0.110  0.096  0.055  0.002  0.000 -0.002 -0.002 -0.003 -0.004 
G8 eliminated. 8 remaining groups of variables.	 Error = 0.14 


Survival indexes : 1 2 3 4 5 6 7 9 
    G3     G2     G1     G7     G5     G6     G9     G4 
 0.099  0.086  0.061  0.001  0.000  0.000 -0.001 -0.004 
G4 eliminated. 7 remaining groups of variables.	 Error = 0.12 


Survival indexes : 1 2 3 5 6 7 9 
    G3     G2     G1     G7     G6     G9     G5 
 0.110  0.092  0.055 -0.001 -0.002 -0.003 -0.005 
G5 eliminated. 6 remaining groups of variables.	 Error = 0.13 


Survival indexes : 1 2 3 6 7 9 
    G3     G2     G1     G7     G6     G9 
 0.118  0.103  0.058  0.003  0.001 -0.003 
G9 eliminated. 5 remaining groups of variables.	 Error = 0.14 


Survival indexes : 1 2 3 6 7 
   G3    G2    G1    G7    G6 
0.131 0.103 0.063 0.006 0.003 
G6 eliminated. 4 remaining groups of variables.	 Error = 0.13 


Survival indexes : 1 2 3 7 
   G3    G2    G1    G7 
0.146 0.120 0.060 0.004 
G7 eliminated. 3 remaining groups of variables.	 Error = 0.14 


Survival indexes : 1 2 3 
   G3    G2    G1 
0.134 0.106 0.071 
G1 eliminated. 2 remaining groups of variables.	 Error = 0.12 


Survival indexes : 2 3 
   G2    G3 
0.183 0.151 
G3 eliminated. 1 remaining groups of variables.	 Error = 0.2 


Survival indexes : 2 
G2 eliminated. No remaining groups of variables.	 Error = 0.44 


Ending...
 3 selected variables:
 G2 G3 G1 



 --- 				 ---
 --- 	Summary functional RFE	 ---
 --- 				 ---


Number of selected variables using a validation set: 3 

Selected variables:
G2 G3 G1 

Validation error for the best model: 0.12 

>   plot(varSel)
> 
>   detach(toyClassif)
> 
> 
> 
> 
> 
> 
> dev.off()
null device 
          1 
>