component.selection: Tune the number of components

In bootsPLS: Bootstrap Subsamplings of Sparse Partial Least Squares - Discriminant Analysis for Classification and Signature Identification

Description

Performs a multiple hypotheses testing procedure to choose the number of components in the splsda analysis

Usage

component.selection(object,alpha,showProgress=TRUE)

Arguments

object	a ‘bootsPLS’ object, as obtained from bootsPLS
alpha	Level of the test
showProgress	Logical. If TRUE, shows the progress of the algorithm.

Details

The testing procedure evaluates the gain in classification accuracy when a new PLS-component is added. This is done by on-sided t-test of level alpha applied on the classification accuracy obtained in the ‘bootsPLS’ object. See the reference below for more details on the multiple testing procedure.

Value

A 'component.selection' object is returned for which plot is available.

pval	pvalue obtained from the testing procedure.
opt	Number of components chosen by the procedure
object	input ‘bootsPLS’ object
alpha	input level of the test

References

Rohart et al. (2016). A Molecular Classification of Human Mesenchymal Stromal Cells. PeerJ, DOI 10.7717/peerj.1845

See Also

plot.component.selection, fit.model, prediction

Examples

## Not run: 
data(MSC)
X=MSC$X
Y=MSC$Y
dim(X)
table(Y)

boot=bootsPLS(X=X,Y=Y,ncomp=3,many=5,kCV=5)

comp=component.selection(boot)

## End(Not run)

bootsPLS documentation built on May 2, 2019, 2:44 a.m.