https://github.com/cran/RandomFields
Tip revision: 6eca414de4c835af2032db4cae6c05e9cc684529 authored by Martin Schlather on 23 April 2016, 15:04:07 UTC
version 3.1.11
version 3.1.11
Tip revision: 6eca414
RFlinearpart.Rd
\name{RFlinearpart}
\alias{RFlinearpart}
\title{Linear part of \command{\link{RMmodel}}}
\description{
\command{\link{RFlinearpart}} returns the linear part of a model
}
\usage{
RFlinearpart(model, x, y = NULL, z = NULL, T = NULL, grid,
data, distances, dim, set=0, ...)
}
\arguments{
\item{model}{object of class \code{\link[=RMmodel-class]{RMmodel}};
the covariance or variogram model, which is to be evaluated}
\item{x}{vector or \eqn{(n \times \code{dim})}{(n x
\code{dim})}-matrix, where \eqn{n} is the number of points at
which the covariance function is to be evaluated;
in particular,
if the model is isotropic or \code{dim=1} then \code{x}
is a vector. \code{x}}
\item{y}{second vector or matrix for non-stationary covariance
functions}
\item{z}{z-component of point if xyzT-specification of points is used}
\item{T}{T-component of point if xyzT-specification of points is used}
\item{grid}{boolean; whether xyzT specify a grid}
\item{data}{vector or matrix of values measured at \code{coord};
If a matrix is given then the columns are interpreted as independent
realisations.\cr
If also a time component is given, then in the data the indices for
the spatial components run the fastest.
If an \code{m}-variate model is used, then each realisation is given as
\code{m} consecutive columns of \code{data}.
}
\item{distances}{vector;
the lower triangular part of the distance matrix column-wise;
equivalently the upper triangular part of the distance matrix row-wise;
either \code{x} or \code{distances} must be missing}
\item{dim}{dimension of the coordinate space in which the model is
applied; only necesary for given \code{distances}}
\item{set}{integer. See section Value for details.}
\item{...}{for advanced
further options and control arguments for the simulation
that are passed to and processed by \command{\link{RFoptions}}
}
}
\note{
In the linear part of the model specification the parameters
that are NA must be the first model part. I.e.
\code{NA * sin(R.p(new="isotropic")) + NA + R.p(new="isotropic")}
is OK, but not
\code{sin(R.p(new="isotropic")) * NA + NA + R.p(new="isotropic")}
}
\value{
\command{\link{RFlinearpart}} returns a list
of three components, \code{Y}, \code{X}, \code{vdim} returning
the deterministic trend, the design matrix, and the multivariability,
respectively.
If \code{set} is positive, \code{Y} and \code{X} contain
the values for the \code{set}-th set of coordinates.
Else, \code{Y} and \code{X} are both lists containing
the values for all the sets.
}
\author{Martin Schlather, \email{schlather@math.uni-mannheim.de}
\url{http://ms.math.uni-mannheim.de/de/publications/software}
}
\seealso{
\link{Bayesian},
\command{\link{RMmodel}},
\command{\link{RFsimulate}},
\command{\link{RFlikelihood}}.
}
\examples{
RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set
## RFoptions(seed=NA) to make them all random again
x <- seq(0, pi, len=10)
trend <- 2 * sin(R.p(new="isotropic")) + 3
model <- RMexp(var=2, scale=1) + trend
print(RFlinearpart(model, x=x)) ## only a deterministic part
trend <- NA * sin(R.p(new="isotropic")) + NA + R.p(new="isotropic") / pi
model <- RMexp(var=NA, scale=NA) + trend
print(RFlinearpart(model, x=x))
\dontshow{FinalizeExample()}
}
\keyword{spatial}