R: Evaluate the robustness of a functional clustering by...

fboot

R Documentation

Evaluate the robustness of a functional clustering by bootstrapping from 1 to (all-1) observations

Description

Evaluate by bootstrapping the robustness of a functional clustering to perturbations of data. The perturbed data can be the number of assemblages taken into account, or the number of performances taken into account.

Usage

fboot(fres,
      opt.var = c("assemblages", "performances"), nbIter = 1,
      opt.nbMax = fres$nbOpt, opt.R2 = FALSE, opt.plot = FALSE,
      filename = "" )

Arguments

`fres`	an object resulting from a functional clustering obtained using the function `fclust`.
`opt.var`	a string, that indicates the variable to test. The option can be `"assemblages"` or `"performances"`.
`nbIter`	an integer, that indicates the number of random drawing to do.
`opt.nbMax`	a logical. If `opt.plot = TRUE`, the trees resulting from leaving out each performance is plotted.
`opt.R2`	a logical. If `opt.R2 = TRUE`, the primary tree is validated and the vectors of coefficient of determination (`R^2`) and efficiency (`E`) are computed.
`opt.plot`	a logical. If `opt.plot = TRUE`, the primary trees resulting from leaving out each performance are plotted. If `opt.R2 = TRUE`, the secondary trees resulting from leaving out each performance are plotted.
`filename`	a string, used as radical for naming the file `"filename.components.csv"`, `"filemane.assemblages.csv"` or `"filemane.performances.csv"` according to the dimensions of matrices.

Details

The trees obtained by bootstrapping of performances to omit are compared to the reference tree obtained with all components using different criteria : "Czekanowski_Dice", "Folkes_Mallows", "Jaccard", "Kulczynski", "Precision", "Rand", "Recall", "Rogers_Tanimoto", "Russel_Rao", "Sokal_Sneath1" and "Sokal_Sneath2" index. For more informations, see the notice of R-package clusterCrit.

Value

a list of lists, each containing a matrix by clustering index.

References

Package "clusterCrit": Clustering Indices, by Bernard Desgraupes (University of Paris Ouest - Lab Modal'X)

Examples


# Enable the comments
oldOption <- getOption("verbose")
if (!oldOption) options(verbose = TRUE)
layout(matrix(c(1,2,3,4), nrow = 2, ncol = 2, byrow = TRUE))




test.boot <- fboot(fres = CedarCreek.2004.2006.res,
                   opt.var  = "performances",
                   nbIter   = 4,
                   opt.plot = TRUE)


layout(1)
options(verbose = oldOption)