Autocorrelation Plot of MCMC Output

Share:

Description

Creates an autocorrelation or partial autocorrelation plot of MCMC output.

Usage

1
2
3
autplot1(x, chain = 1, lag.max = NULL, partial = FALSE,
col = mcmcplotsPalette(1), style = c("gray", "plain"),
ylim = NULL, ...)

Arguments

x

an mcmc.list object with a single variable.

chain

the number of the parallel chain for plotting. The default is to use the first parallel chain.

lag.max

passed as an argument to the autocorrelation function acf.

partial

logical indicating whether paritial autocorrelation should be plotted.

col

color of the bars in the plot.

style

if "gray", then the plotting region is printed with a gray background, otherwise the default plotting region is used.

ylim

limits for the y-axis.

...

further arguments passed to the plotting function.

Details

None.

Value

Creates a plot.

Author(s)

S. McKay Curtis (adapted from Martyn Plummer's autcorr.plot code in the coda package)

References

None.

See Also

acf, autocorr.plot

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
## Create fake MCMC output
nc <- 10; nr <- 1000
pnames <- c(paste("alpha[", 1:5, "]", sep=""), paste("gamma[", 1:5, "]", sep=""))
means <- rpois(10, 20)
fakemcmc <-
    coda::as.mcmc.list(
        lapply(1:3,
               function(i)
                   coda::mcmc(matrix(rnorm(nc*nr, rep(means,each=nr)),
                                     nrow=nr, dimnames=list(NULL,pnames)))))

autplot1(fakemcmc[, "alpha[1]", drop=FALSE])
autplot1(fakemcmc[, "alpha[1]", drop=FALSE], chain=2, style="plain")
autplot1(fakemcmc[, "alpha[1]", drop=FALSE], partial=TRUE)