compute_mu_s: Compute posterior mean and precision of s

In templateICAr: Estimate Brain Networks and Connectivity with ICA and Empirical Priors

Compute posterior mean and precision of s

Description

Compute posterior mean and precision matrix of s

Usage

compute_mu_s(y_vec, D, Dinv_s0, R_inv, theta, P, C_diag)

Arguments

y_vec	Vectorized, dimension-reduced fMRI data, grouped by locations. A vector of length QV.
D	Sparse diagonal matrix of template standard deviations
Dinv_s0	The inverse of D times s0_vec
R_inv	Estimate of inverse spatial correlation matrix (sparse)
theta	List of current parameter estimates
P	Permutation matrix for regrouping by locations (instead of by ICs.)
C_diag	Diagonals of residual covariance of the first level model. A vector of length Q.

Value

A list containing the posterior mean \mu (mu) and precision \Omega (Omega) of s=(s1,...,sQ), along with the supporting vector m, where \mu = \Omega^{-1}m.

templateICAr documentation built on April 3, 2025, 7:41 p.m.