Markov Chain Monte Carlo Objects

Usage

mcmc(data= NA, start = 1, end = numeric(0), thin = 1, ts.eps = .Options$ts.eps))

Arguments

data a vector, matrix or 3d-array of MCMC output
start the iteration number of the first observation
end the iteration number of the last observation
thin the thinning interval between consecutive observations
ts.eps Hmmm. What's this doing here? (MTP)

Description

The function `mcmc' is used to create Markov Chain Monte Carlo objects. These are three dimensional arrays with a class of `ts' (and additional attributes) which represent data which has been produced by a discrete time Markov chain.

MCMC objects resemble time series (ts) objects, but have the additional property that the Markov Chain may be replicated, using a different starting value and/or different seed for the pseudo random number generator. Hence mcmc objects are always three dimensional arrays, with the first dimension representing time, the second dimension representing variables and the third dimension representing replicates of the Markov chain.

start and end{end} are integers which correspond to iteration numbers.

as.mcmc and is.mcmc respectively coerce a vector, matrix or array into a Markov Chain and test whether the object is a Markov Chain.

Markov Chains have methods associated with the generic print, start, end, thin, time and summary functions.

Note

When subscripting mcmc objects, the default value of the drop parameter is FALSE. Hence mcmc objects will always retain their dimensionality and their class when subscripted unless you explicitly specify drop=TRUE.

Author(s)

Martyn Plummer

See Also

start, end, frequency, thin, time, window, join.mcmc.


[Package Contents]