BayesGLM2: Group-level Bayesian GLM

Description

Performs group-level Bayesian GLM estimation and inference using the joint approach described in Mejia et al. (2020).

Usage

BayesGLM2(
  results,
  contrasts = NULL,
  quantiles = NULL,
  excursion_type = NULL,
  contrast_names = NULL,
  gamma = 0,
  alpha = 0.05,
  nsamp_theta = 50,
  nsamp_beta = 100,
  num_cores = NULL,
  verbose = 1
)
BayesGLM_group(
  results,
  contrasts = NULL,
  quantiles = NULL,
  excursion_type = NULL,
  gamma = 0,
  alpha = 0.05,
  nsamp_theta = 50,
  nsamp_beta = 100,
  num_cores = NULL,
  verbose = 1
)

Value

A list containing the estimates, PPMs and areas of activation for each contrast.

Arguments

results

Either (1) a length \(N\) list of "BayesGLM" objects, or (2) a length \(N\) character vector of files storing "BayesGLM" objects saved with saveRDS.

contrasts

(Optional) A list of contrast vectors that specify the group-level summaries of interest. If NULL, use contrasts that compute the average of each field (task HRF) across subjects and sessions.

Each contrast vector is length \(K * S * N\) vector specifying a group-level summary of interest, where \(K\) is the number of fields (task HRFs), \(S\) is the number of sessions, and \(N\) is the number of subjects. For a single subject-session the contrast for the first field would be:

contrast1 <- c(1, rep(0, K-1))

and so the full contrast vector representing the group average across sessions and subjects for the first task would be:

rep(rep(contrast1, S), N) /S /N.

To obtain the group average for the first task, for just the first sessions from each subject:

rep(c(contrast1, rep(0, K*(S-1))), N) /N.

To obtain the mean difference between the first and second sessions, for the first task:

rep(c(contrast1, -contrast1, rep(0, K-2)), N) /N.

To obtain the mean across sessions of the first task, just for the first subject:

c(rep(contrast1, S-1), rep(0, K*S*(N-1)) /S.

quantiles

(Optional) Vector of posterior quantiles to return in addition to the posterior mean.

excursion_type

(For inference only) The type of excursion function for the contrast (">", "<", "!="), or a vector thereof (each element corresponding to one contrast). If NULL, no inference performed.

contrast_names

(Optional) Names of contrasts.

gamma

(For inference only) Activation threshold for the excursion set, or a vector thereof (each element corresponding to one contrast). Default: 0.

alpha

(For inference only) Significance level for activation for the excursion set, or a vector thereof (each element corresponding to one contrast). Default: .05.

nsamp_theta

Number of theta values to sample from posterior. Default: 50.

nsamp_beta

Number of beta vectors to sample conditional on each theta value sampled. Default: 100.

num_cores

The number of cores to use for sampling betas in parallel. If NULL (default), do not run in parallel.

verbose

Should updates be printed? Use 1 (default) for occasional updates, 2 for occasional updates as well as running INLA in verbose mode (if applicable), or 0 for no updates.

INLA Requirement

This function requires the INLA package, which is not a CRAN package. See https://www.r-inla.org/download-install for easy installation instructions.