lme_mass_rfx: Estimation of subject-specific random effects estimates at...

In Deep-MI/fslmer: Univariate And Mass-Univariate Linear Mixed-Effects Analysis For FreeSurfer Imaging Data

Estimation of subject-specific random effects estimates at each vertex

Description

Estimation of subject-specific random effects estimates at each vertex

Usage

lme_mass_rfx(stats, X, Zcols, Y, ni, maskvtx = NA, prs = 1)

Arguments

stats	Structure array containing statistics for every voxel/vertex (generated with either lme_mass_fit_Rgw or lme_mass_fit_vw)
X	Ordered design matrix (according to time for each subject)
Zcols	Vector with the indices of the colums of X that are considered as random effects
Y	Ordered data matrix (n x nv, n=total number of scans and nv=number of vertices/voxels)
ni	Vector whose m entries are the number of repeated measures for each subject (ordered according to X)
maskvtx	Mask's vertices (1-based). Default: NA (all vertices included)
prs	Number of cores for parallel computing (default: 1)

Value

This function returns the subject-specific random effects estimates at each vertex. The output is a list of lists, with the following entries: Rfx: Estimated subject-specific random effects matrix (m x nrfx*nv). The columns of this matrix are grouped by vertex. For example if there are two random effects in the model then the first two columns contain the subject-specific random effect coefficients for the first vertex, then the next two columns contain the subject-specific random effect coefficients for the second vertex and so on ... nrfx: Number of random effects (length(Zcols)). Bhat: Population-level regression coefficients in stats stacked in one matrix.

Examples

## Not run: fitRgw <- lme_mass_fit_Rgw(X, Zcols, Y, ni, fitInit$Theta0, RgGrow$Rgs, Surf)
## Not run: rfx <- lme_mass_rfx(fitRgw$stats, X, Zcols, Y, ni, maskvtx)

Deep-MI/fslmer documentation built on Jan. 24, 2025, 11:24 p.m.