\name{CheckXT} \alias{CheckXT} \alias{PrepareModel} \alias{convert.to.readable} \alias{plotWithCircles} \alias{GetDistributionNames} \title{Internal functions -- do not use them directly} \description{ \code{CheckXT} checks whether the coordinates of the data and related parameters are specified correctly and transforms the coordinates into a standard format \code{PrepareModel} checks whether the parameters of the covariance model and related parameters are specified correctly and transforms the parameters into a standard format \code{convert.to.readable} is the inverse function to \code{PrepareModel}; see Details \code{plotWithCircles} displays data values of marked point processes by circles \code{GetDistributionNames} returns the names of the currently available marginal distributions of the random fields } \usage{ CheckXT(x, y, z, T, grid, gridtriple) PrepareModel(model, param, timespacedim, trend, method=NULL, named=FALSE) convert.to.readable(l, allowed=c("standard", "nested", "list")) plotWithCircles(data, factor=1.0, xlim=range(data[,1])+c(-maxr,maxr), ylim=range(data[,2])+c(-maxr,maxr),col=1, fill=0, ...) GetDistributionNames() } \arguments{ \item{x}{\code{x} coordinates} \item{y}{\code{y} coordinates} \item{z}{\code{z} coordinates} \item{T}{time instances} \item{grid}{see \command{\link{GaussRF}}} \item{gridtriple}{see \command{\link{GaussRF}}} \cr \item{model}{see \command{\link{GaussRF}}} \item{param}{see \command{\link{GaussRF}}} \item{timespacedim}{dimension of the random field including the time dimension, if there is any} \item{trend}{mean or trend of the random field} \item{method}{simulation method} \item{named}{logical. If \code{TRUE} \code{covnr} and \code{param} are returned with names} \cr \item{l}{list as returned by \code{PrepareModel}} \item{allowed}{allowed output formats, see \command{\link{CovarianceFct}}} \cr \item{data}{matrix of 3 columns; first two columns give the coordinates, the third the data} \item{factor}{enlargement factor for data} \item{xlim}{see \command{\link[graphics]{plot}}} \item{ylim}{see \command{\link[graphics]{plot}}} \item{col}{border colour of circles} \item{fill}{filling colour of circles} \item{...}{further graphical parameters} } \details{ \code{convert.to.readable} is roughly speaking the inverse function to \code{PrepareModel}. \code{convert.to.readable} also tries to simplify the model definition, but cannot rediscover the given method for the simulation of the nugget effect in all cases. Due to the simplification in \code{convert.to.readable} and the special definition of the nugget effect for nested models, \code{convert.to.readable} may return a correct model definition in case of incorrect input, namely if \code{scale} is set to \eqn{0} in a list definition, see Examples. } %\value{ % lists of internal parameters %} \author{Martin Schlather, \email{schlath@hsu-hh.de} \url{http://www.unibw-hamburg.de/WWEB/math/schlath/schlather.html}} \seealso{\command{\link{CovarianceFct}}} \keyword{ spatial }%-- one or more ... \keyword{internal} \examples{ % library(RandomFields) x <- function(...) { str(PrepareModel(...)) cat("--------------------------------\n") str(convert.to.readable(PrepareModel(...))) } model <- list(list(model="whi", kappa=5, var=2, s=4), "+", list(model="whi", kappa=1, var=3, s=0)) ## s=0 should not be used only in ## a model definition where the parameters are ## are given in a matrix, see the result x(model=model, ti=1, me="ci") ## since convert.to.readable performs a one-step simplification, ## iterative calls may further simplify the model xx <- convert.to.readable(PrepareModel(model=model, ti=1, me="ci")) x(model=xx$mo, pa=xx$pa, ti=1, me=xx$me) ## back to the matrix definition of nested models str(convert.to.readable(PrepareModel(xx, ti=1), allowed="nested")) ## back to the (correct) list definition str(convert.to.readable(PrepareModel(xx, ti=1), allowed="list")) }