| randboot.multiblock {ade4} | R Documentation |
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
## S3 method for class 'multiblock' randboot(object, nrepet = 199, optdim, ...)
object |
|
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 |
A list containing objects of class krandboot
Stephanie Bougeard (stephanie.bougeard@anses.fr) and Stephane Dray (stephane.dray@univ-lyon1.fr)
Carpenter, J. and Bithell, J. (2000) Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians.Statistics in medicine, 19, 1141-1164
mbpcaiv, mbpls,
testdim.multiblock, as.krandboot
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)