parcorplot {mcmcplots}

Correlation Plot for MCMC Draws of Model Parameters

Description

Creates an image plot of posterior correlations between model parameters from an MCMC simulation.

Usage

parcorplot(mcmcout, parms = NULL, regex = NULL, random = NULL, col = gray(11:0/11), ...)

Arguments

mcmcout

posterior draws. This argument will be coerced to an mcmc object.

parms

character vector specifying which subsets of parameters to plot. If NULL and regex=NULL, mcmcplot will plot all parameters. Regular expressions are used to strip all numbers and punctuation out of the parameter names to find the parameters that match the character strings in parms.

regex

character vector of regular expressions denoting groups of parameters to plot.

random

integer specifying how many parameters from each group will be randomly selected for plotting. This argument is useful when mcmcout has a large number of parameters (e.g., from a hierarchical model). If NULL, mcmcplot will plot all parameters.

col

colors to be used in the plot.

...

further arguments that are passed to the corplot function.

Details

The parcorplot is a wrapper function to corplot that allows the use of arguments parms, regex, and random to conveniently select parameters from an MCMC simulation to plot with corplot.

Value

Creates a plot.

Author(s)

S. McKay Curtis

See Also

corplot, parms2plot

Examples

Rho1 <- outer(1:10, 1:10, function(i, j) 0.5^(abs(i-j)))
Rho2 <- outer(1:5, 1:5, function(i, j) 0.25^(i!=j))
dat1 <- t(apply(matrix(rnorm(10*1000), 1000, 10), 1, function(z, Rho1) t(Rho1)%*%z, Rho1))
dat2 <- t(apply(matrix(rnorm(5*1000), 1000, 5), 1, function(z, Rho2) t(Rho2)%*%z, Rho2))
colnames(dat1) <- paste("theta[", 1:10, "]", sep="")
colnames(dat2) <- paste("alpha[", 1:5, "]", sep="")
dat <- cbind(dat1, dat2)
parcorplot(dat, "theta", col=gray(31:0/31), cex.axis=0.75)
parcorplot(dat, col=heat.colors(31), cex.axis=0.75)
parcorplot(dat, col=topo.colors(31), cex.axis=0.75)
parcorplot(dat, col=terrain.colors(31), cex.axis=0.75)
parcorplot(dat, col=cm.colors(31), cex.axis=0.75)