GLMEM_objfn: Objective function for the BayesGLM EM algorithm

In BayesfMRI: Spatial Bayesian Methods for Task Functional MRI Studies

Objective function for the BayesGLM EM algorithm

Description

This returns the negative of the expected log-likelihood function.

Usage

GLMEM_objfn(theta, spde, model_data, Psi, K, A, n_threads = NULL, Ns = NULL)

Arguments

theta	a vector containing kappa2, phi, and sigma2, in that order
spde	the spde object
model_data	the model_data object containing y and X
Psi	a conversion matrix (N by V) (or N by n)
K	number of covariates
A	The value for Matrix::crossprod(X%*%Psi) (saves time on computation)
n_threads	Needed for SQUAREM (it is an argument to the fixed-point functions)
Ns	The number of samples used to approximate traces using the Hutchinson estimator. If set to 0, the exact trace is found.

Value

A scalar value for the negative expected log-likelihood

BayesfMRI documentation built on April 4, 2025, 1:58 a.m.