MCMC proposal distributions

MCMC proposal functions

mvn.rw

mvn.diag.rw

mvn.rw.adaptive


    square numeric matrix with row- and column-names.
    Specifies the variance-covariance matrix for a multivariate normal random-walk proposal distribution.
  

rw.var


    named numeric vector;
    random-walk SDs for a multivariate normal random-walk proposal with diagonal variance-covariance matrix.
  

rw.sd


    parameters to control the proposal adaptation algorithm.
    Beginning with MCMC iteration <code>scale.start</code>, the scale of the proposal covariance matrix will be adjusted in an effort to match the <code>target</code> acceptance ratio.
    The parameters <code>scale.cooling</code> and <code>max.scaling</code> adjust the scale adaptation.
    Beginning with iteration <code>shape.start</code>, a scaled empirical covariance matrix will be used for the proposals.
  

scale.start, scale.cooling, shape.start, target, max.scaling


  Functions to construct proposal distributions for use with MCMC methods.


programming

Tools for working with partially observed Markov process (POMP) models (also known as stochastic dynamical systems, hidden Markov models, and nonlinear, non-Gaussian, state-space models).  The package provides facilities for implementing POMP models, simulating them, and fitting them to time series data by a variety of frequentist and Bayesian methods.  It is also a versatile platform for implementation of inference methods for general POMP models.

MCMC proposal distributions: MCMC proposal distributions

Description

Usage

Arguments

Value

See Also