mfa: multiple factor analysis

In bbuchsbaum/neuroca: neuroca: multiblock component analysis for neuroimaging data

multiple factor analysis

Description

mfa

Usage

mfa(
  X,
  ncomp = 2,
  preproc = center(),
  normalization = c("MFA", "RV", "None", "RV-MFA", "Dual-RV", "Frob", "custom"),
  M = NULL,
  A = NULL,
  ...
)

Arguments

X	a block_matrix object
ncomp	the number of components to estimate
preproc	a preprocessing pipeline, default is 'center()'
normalization	the normalization method: MFA, RV, RV-MFA, or None (see details).
M	custom weight matrix for the rows
A	custom weight matrix for the columns

Examples

X <- block_matrix(replicate(5, { matrix(rnorm(10*10), 10, 10) }, simplify=FALSE))
res <- mfa(X, ncomp=3, normalization="MFA")
p <- partial_scores(res, 1)
stopifnot(ncol(scores(res)) == 3)

labs <- letters[1:10]
cfier <- classifier(res, labels=labs, ncomp=3)
pred <- predict(cfier, X[1:2,])
cfier2 <- classifier(res, labels=labs, ncomp=3, colind=res$block_indices[[2]])
pred2 <- predict(cfier2, X[1:2,res$block_indices[[2]]])

bbuchsbaum/neuroca documentation built on April 22, 2022, 2:50 a.m.