disco: Distinctive and Common Components with SCA - DISCO

In multiblock: Multiblock Data Fusion in Statistics and Machine Learning

Distinctive and Common Components with SCA - DISCO

Description

This is a wrapper for the DISCOsca function by Zhengguo Gu for computing DISCO.

Usage

disco(X, ncomp = 2, ...)

Arguments

X	list of input blocks.
ncomp	integer number of components to extract.
...	additional arguments (not used).

Details

DISCO is a restriction of SCA where Alternating Least Squares is used for estimation of loadings and scores. The SCA solution is rotated towards loadings (in sample linked mode) which are filled with zeros in a pattern resembling distinct, local and common components. When used in sample linked mode and only selecting distinct components, it shares a resemblance to SO-PLS, only in an unsupervised setting. Explained variances are computed as proportion of block variation explained by scores*loadings'.

Value

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

References

Schouteden, M., Van Deun, K., Wilderjans, T. F., & Van Mechelen, I. (2014). Performing DISCO-SCA to search for distinctive and common information in linked data. Behavior research methods, 46(2), 576-587.

See Also

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

Examples

data(potato)
potList <- as.list(potato[c(1,2,9)])
pot.disco  <- disco(potList)
plot(scores(pot.disco), labels="names")

multiblock documentation built on Nov. 18, 2023, 5:06 p.m.