parallel_fusion_SMC_mixG: Generalised Monte Carlo Fusion [parallel]

In rchan26/hierarchicalFusion: Divide-and-Conquer Monte Carlo Fusion

Generalised Monte Carlo Fusion [parallel]

Description

Generalised Monte Carlo Fusion with mixture Gaussian target

Usage

parallel_fusion_SMC_mixG(
  particles_to_fuse,
  N,
  m,
  time,
  n_comp,
  weights,
  means,
  sds,
  betas,
  precondition_values,
  bounds_multiplier = 1.1,
  resampling_method = "multi",
  ESS_threshold = 0.5,
  seed = NULL,
  n_cores = parallel::detectCores()
)

Arguments

particles_to_fuse	list of length m, where particles_to_fuse[[c]] contains the particles for the c-th sub-posterior (a list of particles to fuse can be initialised by initialise_particle_sets() function)
N	number of samples
m	number of sub-posteriors to combine
time	time T for fusion algorithm
n_comp	integer number of components of mixture Gaussian
weights	vector: weights of mixture Gaussian
means	vector: means of mixture Gaussian
sds	vector: st.devs of mixture Gaussian
betas	vector of length c, where betas[c] is the inverse temperature value for c-th posterior
precondition_values	vector of length m, where precondition_values[c] is the precondition value for sub-posterior c
bounds_multiplier	scalar value to multiply bounds by (should greater than or equal to 1)
resampling_method	method to be used in resampling, default is multinomial resampling ('multi'). Other choices are stratified resampling ('strat'), systematic resampling ('system'), residual resampling ('resid')
ESS_threshold	number between 0 and 1 defining the proportion of the number of samples that ESS needs to be lower than for resampling (i.e. resampling is carried out only when ESS < N*ESS_threshold)
seed	seed number - default is NULL, meaning there is no seed
n_cores	number of cores to use

Value

A list with components:

particles

particles returned from fusion sampler

proposed_samples

proposal samples from fusion sampler

time

run-time of fusion sampler

ESS

effective sample size of the particles after each step

CESS

conditional effective sample size of the particles after each step

resampled

boolean value to indicate if particles were resampled after each time step

precondition_values

list of length 2 where precondition_values[[2]] are the pre-conditioning values that were used and precondition_values[[1]] are the combined precondition values

rchan26/hierarchicalFusion documentation built on Sept. 11, 2022, 10:30 p.m.