pottshm

This function returns a simulation of a <code>n</code> by <code>m</code> grid
from a Potts distribution with <code>ncol</code> colours and a four neighbour
structure, using a Metropolis-Hastings step that avoids proposing
a value identical to the current state of the Markov chain.

Gibbs sampling

Metropolis-Hastings algorithm

Potts model

Allows the reenactment of the R programs used in
the book Bayesian Essentials with R without further programming.
R code being available as well, they can be modified by the user
to conduct one's own simulations.
Marin J.-M. and Robert C. P. (2014) <doi:10.1007/978-1-4614-8687-9>.

JeanMichel Marin

bayess

Bayesian Essentials with R

pottshm function

<dl> <dt>ncol</dt>
<dd>number of colors</dd> <dt>niter</dt>
<dd>number of Metropolis-Hastings iterations</dd> <dt>n</dt>
<dd>number of rows in the image</dd> <dt>m</dt>
<dd>number of columns in the image</dd> <dt>beta</dt>
<dd>parameter of the Potts model</dd></dl>

Arguments

Metropolis-Hastings sampler for a Potts model with ncol classes — pottshm

<dl>

 <dt>ncol</dt>
<dd>number of colors</dd>

 <dt>niter</dt>
<dd>number of Metropolis-Hastings iterations</dd>

 <dt>n</dt>
<dd>number of rows in the image</dd>

 <dt>m</dt>
<dd>number of columns in the image</dd>

 <dt>beta</dt>
<dd>parameter of the Potts model</dd>

</dl>

pottshm: Metropolis-Hastings sampler for a Potts model with `ncol` classes

Description

Usage

Value

Arguments

See Also

Examples