1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
|
\name{mcarray.object}
\alias{mcarray.object}
\alias{print.mcarray}
\alias{summary.mcarray}
\alias{as.mcmc.list.mcarray}
\title{Objects for representing MCMC output}
\description{
An \code{mcarray} object is used by the \code{jags.samples} function
to represent MCMC output from a JAGS model. It is an array with named
dimensions, for which the dimensions "iteration" and "chain" have a
special status
}
\usage{
\method{summary}{mcarray}(object, FUN, \ldots)
\method{print}{mcarray}(x, \ldots)
\method{as.mcmc.list}{mcarray}(x, \ldots)
}
\arguments{
\item{object,x}{an \code{mcarray} object}
\item{FUN}{a function to be used to generate summary statistics}
\item{\ldots}{additional arguments to the call}
}
\value{
The \code{summary} method for \code{mcarray} objects applies the
given function to the array, marginalizing the "chain" and "iteration"
dimensions.
The \code{print} method applies the summary function with
\code{FUN=mean}.
The \code{as.mcmc.list} method coerces an \code{mcarray} to an
\code{mcmc.list} object so that the diagnostics provided by the
\code{coda} package can be applied to the MCMC output it represents.
}
\details{
The \code{coda} package defines \code{mcmc} objects for representing
output from an MCMC sampler, and \code{mcmc.list} for representing
output from multiple parallel chains. These objects emphasize the
time-series aspect of the MCMC output, but lose the original array
structure of the variables they represent. The \code{mcarray} class
attempts to rectify this by preserving the dimensions of the original
node array defined in the JAGS model.
}
\author{Martyn Plummer}
\keyword{models}
|
