multiblock (version 0.8.0)

mbrda: Multiblock Redundancy Analysis - mbRDA

Description

This is a wrapper for the ade4::mbpcaiv function for computing mbRDA.

Usage

mbrda(X, Y, ncomp = 1, ...)

Arguments

X

list of input blocks.

Y

matrix of responses.

ncomp

integer number of PLS components.

...

additional arguments to ade4::mbpcaiv.

Value

multiblock, mvr object with scores, block-scores and block-loading. Relevant plotting functions: multiblock_plots and result functions: multiblock_results.

Details

mbRDA is a multiblock formulation of Redundancy (Data) Analysis. RDA is theoretically between PLS and GCA. Like GCA, RDA does not consider correlations within X, but like PLS it does consider correlations within Y. RDA can also be viewed as a PCR of Y constrained to have scores that are also linear combinations of X. If the adegraphics package is attached, a nice overview can be plotted as plot(mbr$mbpcaiv) following the example below.

References

Bougeard, S., Qannari, E.M., Lupo, C., andHanafi, M. (2011). From Multiblock Partial Least Squares to Multiblock Redundancy Analysis. A Continuum Approach. Informatica, 22(1), 11<U+2013>26.

See Also

Overviews of available methods, multiblock, and methods organised by main structure: basic, unsupervised, asca, supervised and complex.

Examples

# NOT RUN {
# Convert data.frame with AsIs objects to list of matrices
data(potato)
potatoList <- lapply(potato, unclass)

mbr <- mbrda(potatoList[c('Chemical','Compression')], potatoList[['Sensory']], ncomp = 10)
print(mbr)
scoreplot(mbr) # Exploiting mvr object structure from pls package
# }