randboot.multiblock: Bootstraped simulations for multiblock methods

In ade4: Analysis of Ecological Data: Exploratory and Euclidean Methods in Environmental Sciences

Bootstraped simulations for multiblock methods

Description

Function to perform bootstraped simulations for multiblock principal component analysis with instrumental variables or multiblock partial least squares, in order to get confidence intervals for some parameters, i.e., regression coefficients, variable and block importances

Usage

## S3 method for class 'multiblock'
randboot(object, nrepet = 199, optdim, ...)

Arguments

object	an object of class multiblock created by mbpls or mbpcaiv
nrepet	integer indicating the number of repetitions
optdim	integer indicating the optimal number of dimensions, i.e., the optimal number of global components to be introduced in the model
...	other arguments to be passed to methods

Value

A list containing objects of class krandboot

Author(s)

Stéphanie Bougeard ([e-mail]) and Stéphane Dray ([e-mail])

References

Carpenter, J. and Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians.Statistics in medicine, 19, 1141-1164.

Bougeard, S. and Dray S. (2018) Supervised Multiblock Analysis in R with the ade4 Package. Journal of Statistical Software, 86 (1), 1-17. doi: 10.18637/jss.v086.i01

See Also

mbpcaiv, mbpls, testdim.multiblock, as.krandboot

Examples

data(chickenk)
Mortality <- chickenk[[1]]
dudiY.chick <- dudi.pca(Mortality, center = TRUE, scale = TRUE, scannf =
FALSE)
ktabX.chick <- ktab.list.df(chickenk[2:5])
resmbpcaiv.chick <- mbpcaiv(dudiY.chick, ktabX.chick, scale = TRUE,
option = "uniform", scannf = FALSE, nf = 4)
## nrepet should be higher for a real analysis
test <- randboot(resmbpcaiv.chick, optdim = 4, nrepet = 10)
test
if(adegraphicsLoaded())
plot(test$bipc) 

ade4 documentation built on Feb. 16, 2023, 7:58 p.m.