scapeMCMC-package {scapeMCMC}R Documentation

MCMC Diagnostic Plots

Description

Markov-chain Monte Carlo diagnostic plots, accompanying the scape package. The purpose of the package is to combine existing tools from the coda and lattice packages, and make it easy to adjust graphical details. It can be useful for anyone using MCMC analysis, regardless of the application.

Details

Import Coleraine MCMC results:
importMCMC traces of likelihoods, parameters, biomass and recruitment
importProj future projections of biomass and catch
Diagnostic plots:
plotTrace trends
plotAuto thinning
plotCumu convergence
plotSplom confounding of parameters
plotDens posterior(s)
plotQuant multiple posteriors on a common y axis
Examples:
xmcmc, xproj MCMC results and projections

The vignette ‘scape/doc/dsc.pdf’ provides background information and introduces the design of scape and scapeMCMC, but is not updated. Additional details are found on the help(package="scapeMCMC") page.

Note

The plot functions assume that MCMC results are stored either as a plain vector (single chain) or in named columns (multiple chains). It should be easy for users to arrange their MCMC results in this way. The examples demonstrate how several data frames can be bound together in nested lists.

The functions Args and ll (package gdata) can be useful for browsing unwieldy functions and objects.

Author(s)

Arni Magnusson arnima@u.washington.edu and Ian Stewart.

See Also

The scape package provides diagnostic plot functions for statistical catch-at-age model fit to data, recommended before starting MCMC analysis. The coda package is a suite of diagnostic functions for MCMC analysis, many of which are used in scapeMCMC.


[Package scapeMCMC version 1.0-3 Index]