boa.chain.gandr {boa}

Gelman and Rubin Convergence Diagnostics

Description

Computes the Gelman and Rubin convergence diagnostics for a list of MCMC sequences. Estimates are calculated from the second half of each sequence.

Usage

boa.chain.gandr(chain, chain.support, alpha, pnames, window, to)

Arguments

chain

List of matrices whose columns and rows contain the monitored parameters and the MCMC iterations, respectively. The iteration numbers and parameter names must be assigned to the dimnames.

chain.support

List of matrices whose columns and rows contain the monitored parameters and the support (lower and upper limits), respectively.

alpha

Quantile (1 - alpha / 2) at which to estimate the upper limit of the shrink factor.

pnames

Character vector giving the names of the parameters to use in the analysis. If omitted, all parameters are used.

window

Proportion of interations to include in the analysis. If omitted, 50% are included.

to

Largest iteration to include in the analysis. If omitted, no upper bound is set.

Value

psrf

A vector containing the Gelman and Rubin (uncorrected) potential scale reduction factors for the monitored parameters.

csrf

A matrix whose columns and rows are the Gelman and Rubin corrected scale reduction factors (i.e. shrink factor estimates at the median and specified quantile of the sampling distribution) and the monitored parameters, respectively. A correction of (df + 3) / (df + 1) is applied to the scale reduction factors.

mpsrf

A numeric value giving the multivariate potential scale reduction factor proposed by Brooks and Gelman.

window

A numeric vector with two elements giving the range of the iterations used in the analysis.

Author(s)

Brian J. Smith, Nicky Best, Kate Cowles

References

1. Brooks, S. and Gelman, A. (1998). General methods for monitoring convergence of iterative simulations. Journal of Computational and Graphical Statistics, 7(4), 434-55.

2. Gelman, A. and Rubin, D. B. (1992). Inference from iterative simulation using multiple sequences. Statistical Science, 7, 457-72.

See Also

boa.plot, boa.plot.bandg, boa.plot.gandr, boa.print.gandr