geoR (version 1.8-1)

lines.variomodel.krige.bayes: Adds a Bayesian Estimate of the Variogram to a Plot

Description

Adds a Bayesian estimate of the variogram model to a plot typically with an empirical variogram. The estimate is a chosen summary (mean, mode or mean) of the posterior distribution returned by the function krige.bayes.

Usage

# S3 method for krige.bayes
lines.variomodel(x, summary.posterior, max.dist, uvec,
                 posterior = c("variogram", "parameters"),  …)

Arguments

x

an object of the class krige.bayes, typically an output of the function krige.bayes.

summary.posterior

specify which summary of the posterior distribution should be used as the parameter estimate. Options are "mean", "median" or "mode". See DETAILS below.

max.dist

numerical, the maximum distance for the x-axis.

uvec

a numerical vector with support points to compute the variogram values. Only used if posterior = "variogram". Defaults to seq(0, max.dist, length = 51).

posterior

indicates whether the the variogram line is based on the posterior of the variogram function (default) or the posterior of the model parameters.

…

arguments passed to the functions lines or curve.

Value

A line with the estimated variogram plot is added to the plot in the current graphics device. No values are returned.

Details

The function krige.bayes returns samples from the posterior distribution of the parameters \((\sigma^2, \phi, \tau^{2}_{rel})\).

This function allows for two basic options to draw a line with a summary of the variogram function.

1. [1.]"variogram": for each sample of the parameters the variogram function is computed at the support points defined in the argument uvec. Then a function provided by the user in the argument summary.posterior is used to compute a summary of the values obtained at each support point.

2. [2.]"parameters": in this case summaries of the posterior distribution of the model parameters as "plugged-in" in the variogram function. One of the options "mode" (default) ,"median" or "mean" can be provided in the argument summary.posterior. The option mode, uses the mode of \((\phi, \tau^{2}_{rel})\) and the mode of of \(\sigma^2\) conditional on the modes of the former parameters. For the options mean and median these summaries are computed from the samples of the posterior.

References

Further information on the package geoR can be found at: http://www.leg.ufpr.br/geoR.

See Also

lines.variomodel, krige.bayes and lines.

Examples

# NOT RUN {
#See examples in the documentation of the function krige.bayes().
# }