fmri (version 1.7-2)

fmriica: Independent Component Analysis

Description

The function performs Independent Component Analysis.

Usage

fmriica(data, m = 3, method = "temporal", xind = NULL, yind =
        NULL, zind = NULL, tind = NULL, filter.time = "None",
        filter.space = FALSE, h.space = 3, h.time = 3, keepv =
        FALSE, ...)

Arguments

data

Observation matrix (dimension Nxd)

m

Number of independent components.

method

Either "spatial" or "temporal". Specifies the type of ICA to perform.

xind

index of x-coordinates to use

yind

index of y-coordinates to use

zind

index of z-coordinates to use

tind

index of time points to use

filter.time

not yet documented

filter.space

not yet documented

h.space

not yet documented

h.time

not yet documented

keepv

not yet documented

…

further arguments to fastICA

Value

The function returns a list with components

ihat

Matrix containing the first m ICA directions as columns.

sdev

Standard deviations of the principal components of the thresholded ICA directions

xhat

first m components of the rotated data

v

If keepv==TRUE the set of directions v^{(k)}

normv

If keepv==TRUE the norm of each v^{(k)}.

Details

This is still experimental code based on the package fastICA. The package fastICA seems limited in the data-size it can handle. xind, yind, zind and tind may be use to restrict the analysis to a cube in space and certain time points.

See Also

ngca