This plot shows the evolution of Gelman and Rubin's shrink factor as the number of iterations increases.
gelman.plot(x, bin.width = 10, max.bins = 50,
confidence = 0.95, transform = FALSE, autoburnin=TRUE, auto.layout = TRUE,
ask, col, lty, xlab, ylab, type, …)an mcmc object
Number of observations per segment, excluding the first segment which always has at least 50 iterations.
Maximum number of bins, excluding the last one.
Coverage probability of confidence interval.
Automatic variable transformation (see gelman.diag)
Remove first half of sequence (see gelman.diag)
If TRUE then, set up own layout for
plots, otherwise use existing one.
Prompt user before displaying each page of plots. Default is
dev.interactive() in R and interactive() in S-PLUS.
graphical parameter (see par)
graphical parameter (see par)
graphical parameter (see par)
graphical parameter (see par)
graphical parameter (see par)
further graphical parameters.
A potential problem with gelman.diag is that it may mis-diagnose
convergence if the shrink factor happens to be close to 1 by chance.
By calculating the shrink factor at several points in time,
gelman.plot shows if the shrink factor has really converged, or
whether it is still fluctuating.
The Markov chain is divided into bins according to the arguments
bin.width and max.bins. Then the Gelman-Rubin shrink factor
is repeatedly calculated. The first shrink factor is calculated with
observations 1:50, the second with observations \(1:(50+bin.width)\),
the third contains samples \(1:(50+2*bin.width)\) and so on.
If the chain has less than \(50 + bin.width\) iterations then
gelman.diag will exit with an error.
Brooks, S P. and Gelman, A. (1998) General Methods for Monitoring Convergence of Iterative Simulations. Journal of Computational and Graphical Statistics, 7, 434-455.