mcmc_updates: Internal functions for MCMC updates

In brieuclehmann/multibergm: Bayesian Exponential Random Graph Models for Multiple Networks

Internal functions for MCMC updates

Description

Functions used internally by to perform Gibbs or exchange updates.

Usage

exchange_update_cp(curr, prop, prior_mean, delta, prior_cov, etamap)

mean_update(obs_data, obs_cov, prior_mean, prior_cov, model_matrix)

cov_update(
  obs_data,
  obs_mean,
  prior_df,
  prior_scale,
  labels = rep(1, nrow(obs_data)),
  curr_cov = NULL
)

ergm_wrapper(coefs, control)

Arguments

curr	A vector or matrix of current mean parameter values.
prop	A vector or matrix of proposed mean parameter values
prior_mean	Prior mean of parameter
delta	Change in summary statistics as produced by ergm_wrapper
prior_cov	Prior covariance of parameter
etamap	The list of values that constitutes the theta-> eta mapping and is returned by ergm.etamap
obs_data	Matrix of observed data
obs_cov	Fixed (known) covariance of observed data
prior_df	Prior degrees of freedom in Inverse-Wishart
labels	Labels to associate each observation with a grouping
curr_cov	Current value of covariance parameter
coefs	Matrix of coefficients to simulate ERGs from
control	A list of parameters set by control_multibergm containing settings for the ERGM simulations
prior_obs_scale	Prior scale matrix in Inverse-Wishart
var_mean	Fixed (known) mean of parameters

Functions

• mean_update: Gibbs update of a mean parameter

• cov_update: Gibbs update of a covariance parameter

• ergm_wrapper: Gibbs update of a mean parameter

brieuclehmann/multibergm documentation built on June 19, 2024, 6:36 p.m.