geoRglm (version 0.9-16)

mcmc.control: Defines options for the MCMC-algorithm

Description

This auxiliary function defines options for the MCMC-algorithm used by pois.krige.bayes, binom.krige.bayes, binom.krige and pois.krige.

Usage

mcmc.control(S.scale, Htrunc, S.start, burn.in, thin, n.iter, phi.start,
             phi.scale)

Arguments

S.scale

proposal variance for the update of \(S\) in the algorithm.

Htrunc

truncation parameter for the MCMC-algorithm. Only used for the Poisson model. Either a number or a vector of the same dimension as the data. Default is Htrunc = 2*data + 5.

S.start

starting value for S (without the mean !!) in the MCMC-algorithm. Default value exist. Should either be a vector of the same length as the data, S.start="random", or S.start="default"

burn.in

length of the burn-in period. Default is 0.

thin

thinning: only every thin scan is stored. Default is thin=10.

n.iter

number of iterations performed. Default is n.iter=1000*thin.

phi.start

starting value for \(\phi\). Default is the median of prior$phi.discrete.

phi.scale

proposal variance for the update of \(\phi\) in the algorithm.

Value

A list with processed arguments to be passed to the main function.

See Also

pois.krige.bayes, binom.krige.bayes, binom.krige and pois.krige.