parameter.plot.table {MIfuns}R Documentation

Tabular and graphic summary of MCMC results

Description

parameter.plot.table creates history, density and Gelman-Rubin-Brooks plots for each parameter in an array of MCMC results and returns a table of summary statistics for the parameters. mcmc.history creates only the history plots, and mcmc.density creates only the density plots.

Usage

        parameter.plot.table(parameter.array)
        mcmc.history(sims.array)
        mcmc.density(sims.array,n=50)

Arguments

parameter.array 3-way array of MCMC simulations with dimensions number of simulations per chain, number of chains, and number of parameters. This is consistent with the parameters.array element of the list object returned by the bugs function of the R2WinBUGS package.
sims.array Output item containing results from WinBUGS fit
n the number of equally spaced points at which the density is to be estimated.

Details

parameter.plot.table requires functions from the lattice and coda packages. See documentation for the coda package for details on estimation of effective sample size and the time-series standard error, and creation of the Gelman-Rubin-Brooks plots. The history, density and Gelman-Rubin-Brooks plots are printed to the current graphics device.

Value

A matrix of summary statistics describing the MCMC simulations for each parameter in sims.array. Each row contains the statistics for one parameter. The columns correspond to the following statistics: mean, standard deviation, naive standard error, time-series standard error, the 2.5th, 25th, 50th, 75th 97.5th percentile, and the effective sample size.

Author(s)

Developed by Metrum Institute (Bill Gillespie and others).

References

MIfuns package available from Metrum Institute website (http://metruminstitute.org/).


[Package MIfuns version 3.2.9 Index]