CPF_coupled: Coupled Conditional Particle Filter

In lolmid/unbiased_intractable_targets: Debiased particle MCMC with couplings

Description

Runs a coupled conditional particle filter, with or without ancestor sampling

Usage

CPF_coupled(nparticles, model, theta, observations, ref_trajectory1,
  ref_trajectory2, coupled_resampling, with_as = FALSE)

Arguments

nparticles	number of particles
model	a list representing a model, for instance as given by get_ar.
theta	a parameter to give to the model functions
observations	a matrix of observations of size datalength x dimension(observation)
ref_trajectory1	a first reference trajectory, of size dimension(process) x datalength
ref_trajectory2	a second reference trajectory, of size dimension(process) x datalength
coupled_resampling	a coupled resampling scheme, such as CR_indexmatching.
with_as	whether ancestor sampling should be used (TRUE/FALSE)

Value

A pair of new trajectories.

lolmid/unbiased_intractable_targets documentation built on May 13, 2019, 11:54 p.m.