redist (version 2.0.2)

redist.diagplot: Diagnostic plotting functionality for MCMC redistricting.

Description

redist.diagplot generates several common MCMC diagnostic plots.

Usage

redist.diagplot(sumstat,
plot = c("trace", "autocorr", "densplot", "mean", "gelmanrubin"),
logit = FALSE, savename = NULL)

Arguments

sumstat

A vector, list, mcmc or mcmc.list object containing a summary statistic of choice.

plot

The type of diagnostic plot to generate: one of "trace", "autocorr", "densplot", "mean", "gelmanrubin". If plot = "gelmanrubin", the input sumstat must be of class mcmc.list or list.

logit

Flag for whether to apply the logistic transformation for the summary statistic. The default is FALSE.

savename

Filename to save the plot. Default is NULL.

Value

Returns a plot of file type .pdf.

Details

This function allows users to generate several standard diagnostic plots from the MCMC literature, as implemented by Plummer et. al (2006). Diagnostic plots implemented include trace plots, autocorrelation plots, density plots, running means, and Gelman-Rubin convergence diagnostics (Gelman & Rubin 1992).

References

Fifield, Benjamin, Michael Higgins, Kosuke Imai and Alexander Tarr. (2016) "A New Automated Redistricting Simulator Using Markov Chain Monte Carlo." Working Paper. Available at http://imai.princeton.edu/research/files/redist.pdf.

Gelman, Andrew and Donald Rubin. (1992) "Inference from iterative simulations using multiple sequences (with discussion)." Statistical Science.

Plummer, Martin, Nicky Best, Kate Cowles and Karen Vines. (2006) "CODA: Convergence Diagnosis and Output Analysis for MCMC." R News.

Examples

# NOT RUN {
data(algdat.pfull)

## Get an initial partition
set.seed(1)
initcds <- algdat.pfull$cdmat[,sample(1:ncol(algdat.pfull$cdmat), 1)]

## 25 precinct, three districts - no pop constraint ##
alg_253 <- redist.mcmc(adjobj = algdat.pfull$adjlist,
popvec = algdat.pfull$precinct.data$pop,
initcds = initcds,nsims = 10000)

## Get Republican Dissimilarity Index from simulations
rep_dmi_253 <- redist.segcalc(alg_253,
algdat.pfull$precinct.data$repvote,
algdat.pfull$precinct.data$pop)

## Generate diagnostic plots
redist.diagplot(rep_dmi_253, plot = "trace")
redist.diagplot(rep_dmi_253, plot = "autocorr")
redist.diagplot(rep_dmi_253, plot = "densplot")
redist.diagplot(rep_dmi_253, plot = "mean")

## Gelman Rubin needs two chains, so we run a second
alg_253_2 <- redist.mcmc(adjobj = algdat.pfull$adjlist,
popvec = algdat.pfull$precinct.data$pop,
initcds = initcds,nsims = 10000)

rep_dmi_253_2 <- redist.segcalc(alg_253_2,
algdat.pfull$precinct.data$repvote,
algdat.pfull$precinct.data$pop)

## Make a list out of the objects:
rep_dmi_253_list <- list(rep_dmi_253, rep_dmi_253_2)


## Generate Gelman Rubin diagnostic plot
redist.diagplot(sumstat = rep_dmi_253_list, plot = 'gelmanrubin')
# }