ess: Code for the elliptical slice sampler

In jtipton25/sgMRA: Multi-Resolution Gaussian Process Approximations with Stochastic Gradient Descent

Code for the elliptical slice sampler

Description

Code for the elliptical slice sampler

Usage

ess(
  current,
  prior,
  prior_mean,
  pars,
  log_like_fun,
  update_pars_fun,
  max_iter = 50
)

Arguments

current	The current state of the parameter of interest in the MCMC algorithm.
prior	A sample from the mean zero, Gaussian prior of the parameter of interest.
prior_mean	The mean, if nonzero, of the Gaussian prior distribution for the parameter of interest.
pars	The set of parameters needed to evaluation the log-likelihood function for the parameter of interest. At a minimum, the pars object must contain the variables parameter which is the parameter to be updated, par_name which is a string name of the variable to update, verbose which is a boolean option for verbose output, and the ess performance tracking variables num_calls_ess, num_max_contract, and num_unusual_contract.
log_like_fun	The log-likelihood function for the parameter of interest.
update_pars_fun	The function to update the parameters pars.
max_iter	The maximum number of iterations to perform for the elliptical slice sampler.

jtipton25/sgMRA documentation built on Feb. 9, 2023, 4:53 a.m.