https://github.com/cran/RandomFields
Tip revision: e994a4415e67fa60cbfd3f208aaab20872521c0b authored by Martin Schlather on 14 February 2019, 21:02:19 UTC
version 3.3
version 3.3
Tip revision: e994a44
RFfit-class.Rd
\name{RFfit-class}
\docType{class}
\alias{RFfit-class}
\alias{RF_fit-class}
\alias{show,RFfit-method}
\alias{persp,RFfit-method}
\alias{print,RFfit-method}
\alias{anova,RFfit-method}
\alias{AIC,RFfit-method}
\alias{BIC,RFfit-method}
\alias{summary,RFfit-methodt}
\alias{[,RFfit-method}
\alias{[,RFfit,ANY,ANY-method}
\alias{[,RFfit,ANY,ANY,ANY-method}
\alias{coerce,RFfit,RFempVariog-method}
\alias{print.RFfit}
\alias{plot,RFfit,missing-method}
\alias{contour.RFfit}
\alias{contour.RFempVariog}
\alias{AICc.RFfit}
\alias{logLik.RFfit}
\alias{print.RF_fit}
\alias{anova.RF_fit}
\alias{AIC.RF_fit}
\alias{BIC.RF_fit}
\alias{AICc.RF_fit}
\alias{summary.RF_fit}
\alias{logLik.RF_fit}
\alias{.RFfit}
\alias{.RF_fit}
\alias{residuals,RFfit-method}
\alias{summary,RFfit-method}
\alias{RFhessian}
%\alias{plot,RFfit-method}
\title{Class \code{RFfit}}
\description{ Class for RandomFields' representation of model estimation
results
}
%anova.RF_fit(object, ...)
%AIC.RF_fit(object, ..., k=2, method="ml", full=TRUE)
%BIC.RF_fit(object, ..., method="ml", full=TRUE)
%summary.RF_fit(object, ..., method="ml", full=FALSE)
%print.RF_fit(x, ..., method="ml", full=FALSE)
%logLik.RF_fit(object, REML = FALSE, ..., method="ml")
\usage{
\S4method{residuals}{RFfit}(object, ..., method="ml", full=FALSE)
\S4method{summary}{RFfit}(object, ..., method="ml")
\S4method{plot}{RFfit,missing}(x, y, ...)
\S3method{contour}{RFfit}(x, ...)
\S3method{contour}{RFempVariog}(x, ...)
RFhessian(model)
}
\arguments{
\item{object}{see the generic function;
}
\item{...}{
\itemize{
\item \command{plot}: arguments to be passed to methods; mainly graphical
arguments, or further models in case of class \code{CLASS_CLIST},
see Details.
\item \command{summary}: see the generic function
\item \command{contour} : see \command{\link{RFplotEmpVariogram}}
}
}
\item{method}{character; only for \code{class(x)=="RFfit"}; a
vector of slot names for which the fitted covariance or variogram
model is to be plotted; should be a subset of
\code{slotNames(x)} for which the corresponding slots are of class
\code{CLASS_FIT}; by default, the maximum likelihood fit
(\code{"ml"}) will be
plotted}
\item{full}{logical.
if \code{TRUE} submodels are reported as well (if available).
}
\item{x}{object of class \code{\link[=RFsp-class]{RFsp}} or
\command{\link[=RFempVariog-class]{RFempVariog}} or
\command{\link[=RFfit-class]{RFfit}} or
\command{\link[=RMmodel-class]{RMmodel}}; in the latter case, \code{x} can
be any sophisticated model but it must be either stationary or a
variogram model}
\item{y}{unused}
\item{model}{
\code{class(x)=="RF_fit"} or \code{class(x)=="RFfit"}, obtained
from \command{\link{RFfit}}
}
}
\details{
for the definition of \command{plot} see \command{\link{RFplotEmpVariogram}}.
}
\section{Creating Objects}{
Objects are created by the function
\command{\link{RFfit}}
}
\section{Slots}{
\describe{
\item{\code{autostart}:}{RMmodelFit; contains the estimation results
for the method 'autostart' including a likelihood value, a constant
trend and the residuals}
\item{\code{boxcox}:}{logical; whether the
parameter of a Box Cox tranformation has been estimated
}
\item{\code{coordunits}:}{string giving the units of the coordinates,
see also option \code{coordunits} of \command{\link{RFoptions}}.
}
\item{\code{deleted}:}{integer vector.
Positions of the parameters that have been deleted to get the set of
variables, used in the optimization.
}
\item{\code{ev}:}{list; list of objects of class
\code{\link[=RFempVariog-class]{RFempVariog}},
contains the empirical variogram estimates of the data}
\item{\code{fixed}:}{
list of two vectors. The fist gives the position where the
parameters are set to zero. The second gives the position where the
parameters are set to one.
}
\item{\code{internal1}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{internal2}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{internal3}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{lowerbounds}:}{RMmodel; covariance model in which each
parameter value gives the lower bound for the respective parameter}
\item{\code{ml}:}{RMmodelFit; analog to slot 'autostart'
}
\item{\code{modelinfo}:}{ table with information on the parameters:
name, boundaries, type of parameter
}
\item{\code{n.covariates}:}{ number of covariates
}
\item{\code{n.param}:}{
number of parameters (given by the user)
}
\item{\code{n.variab}:}{
number of variables (used internally);
\code{n.variab} is always less than or equal to \code{n.param}
}
\item{\code{number.of.data}:}{
the number of data values passed to \command{\link{RFfit}} that are
not \code{NA} or \code{NaN}
}
\item{\code{number.of.parameters}:}{
total number of parameters of the model that had to be estimated
including variances, scales, co-variables, etc.
}
\item{\code{p.proj}:}{vector of integers. The original position of those
parameters that are used in the submodel
}
\item{\code{plain}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{report}:}{
If not empty, it indicates that this model should be reported
and gives a standard name of the model.
Various functions, e.g. \command{print.RMmodelFit}, use
this information if their argument \code{full} equals \code{TRUE}.
}
\item{\code{self}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{sd.inv}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{sqrt.nr}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{submodels}:}{
list. Sequence (in some cases even nested sequence)
of models that is used to determine an initial value in
\command{}
}
\item{\code{table}:}{matrix; summary of estimation results of
different methods}
\item{\code{transform}:}{function; }
\item{\code{true.tsdim}:}{
time space dimension of the (original!) data,
even for submodels that consider parts of separable models.
}
\item{\code{true.vdim}:}{
multivariability of the (original!) data,
even for submodels that consider independent models
for the multivariate components.
}
\item{\code{upperbounds}:}{RMmodel; see slot 'lowerbounds'}
\item{\code{users.guess}:}{RMmodelFit; analog to slot 'autostart'}
\item{\code{ml}:}{RMmodelFit; analog to slot 'autostart'; with maximum
likelihood method}
\item{\code{v.proj}:}{vector of integers.
The components selected in one of the submodels
}
\item{\code{varunits}:}{string giving the units of the variables,
see also option \code{varunits} of \command{\link{RFoptions}}.
}
\item{\code{x.proj}:}{
logical or integer. If logical, it means that no
separable model is considered there. If integer, then
it gives the considered directions of a separable model.
}
\item{\code{Z}:}{
standardized list of information on the data
}
}
}
%\section{Extends}{
%}
\section{Methods}{
\describe{
\item{plot}{\code{signature(x = "RFfit")}: gives a plot of the
empirical variogram together with the fitted model, for more details see
\command{\link{plot-method}}.
}
\item{show}{\code{signature(x = "RFfit")}: returns the structure
of \code{x}
}
\item{persp}{\code{signature(obj =
"RFfit")}: generates \command{\link[graphics]{persp}} plots
}
\item{print}{\code{signature(x = "RFfit")}: identical with
\command{show}-method, additional argument is \code{max.level}
}
\item{[}{\code{signature(x = "RFfit")}: enables accessing
the slots via the \code{"["}-operator, e.g. \code{x["ml"]}
}
\item{as}{\code{signature(x = "RFfit")}:
converts into other formats, only implemented for target class
\code{\link[=RFempVariog-class]{RFempVariog}}
}
\item{anova}{performs a likelihood ratio test base on a chisq approximation
}
\item{summary}{provides a summary}
\item{logLik}{provides an object of class \code{"logLik"}
}
\item{AIC,BIC}{provides the AIC and BIC information, respectively}
\item{\code{signature(x = "RFfit", y = "missing")}}{Combines the plot of
the empirical variogram with the estimated covariance or variogram
model (theoretical) curves; further models can be added via the
argument \code{model}.}
}
}
\section{Further 'methods'}{
\code{AICc.RFfit(object, ..., method="ml", full=FALSE)}
\code{AICc.RF_fit(object, ..., method="ml", full=TRUE)}
}
%\section{Details}{
%}
\author{Alexander Malinowski; \martin}
\seealso{
\code{\link{RFfit}},
\code{\link{RFvariogram}},
\code{\link{RMmodel-class}},
\code{\link{RMmodelFit-class}},
\code{\link{plot-method}}.
}
\references{
AICc:
\itemize{
\item Hurvich, C.M. and Tsai, C.-L. (1989)
Regression and Time Series Model Selection in Small Samples
\emph{Biometrika}, \bold{76}, 297-307.
}
}
\examples{\dontshow{StartExample()}
# see RFfit
\dontshow{FinalizeExample()}
}
\keyword{classes}
\keyword{print}
\keyword{hplot}