multiblock (version 0.8.0)

gca: Generalized Canonical Analysis - GCA

Description

This is an interface to both SVD-based (default) and RGCCA-based GCA (wrapping the RGCCA::rgcca function)

Usage

gca(X, ncomp = 2, svd = TRUE, tol = 10^-12, corrs = TRUE, ...)

Arguments

X

list of input blocks.

ncomp

integer number of components to extract.

svd

logical indicating if Singular Value Decomposition approach should be used (default=TRUE).

tol

numeric tolerance for component inclusion (singular values).

corrs

logical indicating if correlations should be calculated for RGCCA based approach.

...

additional arguments for RGCCA approach.

Value

multiblock object including relevant scores and loadings. Relevant plotting functions: multiblock_plots and result functions: multiblock_results.

Details

GCA is a generalisation of Canonical Correlation Analysis to handle three or more blocks. There are several ways to generalise, and two of these are available through gca. The default is an SVD based approach estimating a common subspace and measuring mean squared correlation to this. An alternative approach is available through RGCCA.

References

• Carroll, J. D. (1968). Generalization of canonical correlation analysis to three or more sets of variables. Proceedings of the American Psychological Association, pages 227-22.

• Van der Burg, E. and Dijksterhuis, G. (1996). Generalised canonical analysis of individual sensory profiles and instrument data, Elsevier, pp. 221<U+2013>258.

See Also

Overviews of available methods, multiblock, and methods organised by main structure: basic, unsupervised, asca, supervised and complex. Common functions for computation and extraction of results and plotting are found in multiblock_results and multiblock_plots, respectively.

Examples

# NOT RUN {
data(potato)
potList <- as.list(potato[c(1,2,9)])
pot.gca <- gca(potList)
plot(scores(pot.gca), labels="names")

# }