mixKernel (version 0.9-2)

plotVar.kernel.pca: Plot importance of variables in kernel PCA

Description

Provides a representation of variable importance in kernel PCA.

Usage

plotVar.kernel.pca(
  object,
  blocks = unique(object$cc.blocks),
  ndisplay = 5,
  ncol = 2,
  ...
)

Arguments

object

: a kernel.pca object returned by kernel.pca.

blocks

a numerical vector indicating the block variables to display.

ndisplay

integer. The number of important variables per blocks shown in the representation. Default: 5.

ncol

integer. Each block of variables is displayed in a separate subfigure. ncol sets the number of columns for the global figure. Default: 2.

...

external arguments.

Author

Jerome Mariette <[e-mail]> Nathalie Vialaneix <[e-mail]>

Details

plotVar.kernel.pca produces a barplot for each block. The variables for which the importance has been computed with kernel.pca.permute are displayed. The representation is limited to the ndisplay most important variables.

References

Crone L. and Crosby D. (1995). Statistical applications of a metric on subspaces to satellite meteorology. Technometrics, 37(3), 324-328.

See Also

kernel.pca, kernel.pca.permute

Examples

data(TARAoceans)

# compute one kernel for the psychem dataset
phychem.kernel <- compute.kernel(TARAoceans$phychem, kernel.func = "linear")
# perform a KPCA
kernel.pca.result <- kernel.pca(phychem.kernel)
# compute importance for all variables in this kernel
kernel.pca.result <- kernel.pca.permute(kernel.pca.result, phychem = colnames(TARAoceans$phychem))

if (FALSE) plotVar.kernel.pca(kernel.pca.result, ndisplay = 10)