ParticleMarginalMH: Run a Particle Marginal Metropolis-Hastings chain

In nickpoison/nltsa: Nonlinear Time Series Analysis

Description

This function implements PMMH by calling a Metropolis-Hastings algorithm on the extended space.

Usage

  ParticleMarginalMH(initial.nlss, y, n.particles, n.mcmc,
    kernel.rnd = random.walk,
    kernel.logpdf = function(from, to) 0,
    prior.logpdf = function(nlss) 0,
    .progress = progress_text, ...)

Arguments

initial.nlss	NLSS whose parameters value will serve as starting point
y	Observations
n.particles	Number of particles
n.mcmc	Number of MCMC iterations
kernel.rnd	Sampling from proposal kernel on the parameter space, default to random.walk
kernel.logpdf	Evaluate the log-pdf of the proposal kernel
prior.logpdf	Evaluate the log-pdf of the prior distribution
.progress	Progress bar to use, from plyr package, default to progress_text
...	Extra parameters to pass to the SISR filter, see function sisr

Value

A list with the same components as MH

See Also

MH, sisr, random.walk

nickpoison/nltsa documentation built on May 23, 2019, 4:48 p.m.