RandomFields: RFspatialGridDataFrame-class – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

RFspatialGridDataFrame-class

Class "RFspatialGridDataFrame"

Description

Class for spatial attributes that have spatial or spatio-temporal locations (at least of dimension 2) on a (full) regular grid. Direct extension of class SpatialGridDataFrame from the sp-package. See sp2RF for an explicit transformation.

Usage

## S4 method for signature 'RFspatialGridDataFrame'
RFspDataFrame2conventional(obj, data.frame=FALSE)

Arguments

`obj`	an `RFspatialGridDataFrame` object
`data.frame`	logical. If `TRUE` a `data.frame` is returned.

Creating Objects

Objects can be created by using the functions RFspatialGridDataFrame or conventional2RFspDataFrame or by calls of the form as(x, "RFspatialGridDataFrame"), where x is of class RFspatialGridDataFrame.

Slots

.RFparams:: list of 2; .RFparams$n is the number of repetitions of the random field contained in the data slot; .RFparams$vdim gives the dimension of the values of the random field, equals 1 in most cases
data:: object of class data.frame; containing attribute data
grid:: object of class GridTopology; grid parameters

bbox:: matrix specifying the bounding box
proj4string:: object of class CRS; projection

Extends

Class "SpatialGridDataFrame", directly. Class "SpatialGrid", by class "SpatialGridDataFrame". Class "Spatial", by class "SpatialGrid".

Methods

contour: signature(obj = "RFspatialGridDataFrame"): generates contour plots
plot: signature(obj = "RFspatialGridDataFrame"): generates nice image plots of the random field; if space-time-dim2, a two-dimensional subspace can be selected using the argument MARGIN; to get different slices in a third direction, the argument MARGIN.slices can be used; for more details see plot-method or type method?plot("RFspatialGridDataFrame")
persp: signature(obj = "RFspatialGridDataFrame"): generates persp plots
show: signature(x = "RFspatialGridDataFrame"): uses the show-method for class SpatialGridDataFrame.
print: signature(x = "RFspatialGridDataFrame"): identical to show-method
RFspDataFrame2conventional: signature(obj = "RFspatialGridDataFrame"): conversion to a list of non-sp-package based objects; the data-slot is converted to an array of dimension [1*(vdim>1) + space-time-dimension + 1*(n>1)]; the grid-slot is converted to a 3-row matrix; the grid definition of a possible time-dimension becomes a separate list element
RFspDataFrame2dataArray: signature(obj = "RFspatialGridDataFrame"): conversion of the data-slot to an array of dimension [space-time-dimension + 2], where the space-time-dimensions run fastest, and vdim and n are the last two dimensions
coordinates: signature(x = "RFspatialGridDataFrame"): calculates the coordinates from grid definition
[: signature(x = "RFspatialGridDataFrame"): selects columns of data-slot; returns an object of class RFspatialGridDataFrame.
[<-: signature(x = "RFspatialGridDataFrame"): replaces columns of data-slot; returns an object of class RFspatialGridDataFrame.
as: signature(x = "RFspatialGridDataFrame"): converts into other formats, only implemented for target class RFspatialPointsDataFrame
cbind: signature(...): if arguments have identical topology, combine their attribute values
range: signature(x = "RFspatialGridDataFrame"): returns the range
hist: signature(x = "RFspatialGridDataFrame"): plots histogram
as.matrix: signature(x = "RFspatialGridDataFrame"): converts data-slot to matrix
as.array: signature(x = "RFspatialGridDataFrame"): converts data-slot to array
as.vector: signature(x = "RFspatialGridDataFrame"): converts data-slot to vector
as.data.frame: signature(x = "RFspatialGridDataFrame"): converts data-slot and coordinates to a data.frame

Details

Note that in the data-slot, each column is ordered according to the ordering of coordinates(grid), the first dimension runs fastest and for all BUT the second dimension, coordinate values are in ascending order. In the second dimension, coordinate values run from high to low. Hence, when converting to conventional formats using RFspDataFrame2conventional or RFspDataFrame2dataArray, the data array is re-ordered such that all dimensions are in ascending order. as.matrix does not perform re-ordering.

Methods summary, and dimensions are defined for the “parent”-class RFsp.

Author(s)

Alexander Malinowski, Martin Schlather, schlather@math.uni-mannheim.de, https://www.wim.uni-mannheim.de/schlather/

Examples

RFoptions(seed=0) ## *ANY* simulation will have the random seed 0; set
##                   RFoptions(seed=NA) to make them all random again

n <- 3

x <- GridTopology(cellcentre.offset=c(0, 0),
 cellsize=c(1, 0.2),
 cells.dim=c(10, 30))
f <- RFsimulate(model=RMexp(), x=x, n=n)

str(f)
str(RFspDataFrame2conventional(f))
str(RFspDataFrame2dataArray(f))
head(coordinates(f))
str(f[2]) ## selects second column of data-slot
all.equal(f, cbind(f,f)[1:3]) ## TRUE
str(as(f, "RFspatialPointsDataFrame"))

plot(f, nmax=2)

steps <- c(10, 1, 10, 10)

x2 <- rbind(c(0, 0, 0, 0),
 c(1, 0.2, 2, 5),
 steps)
scale <- 10

f2 <- RFsimulate(model=RMwhittle(nu=1.2, scale=scale), x=x2, n=n,
                 grid = TRUE)
plot(f2, MARGIN=c(3,4), MARGIN.slices=1, n.slices=6, nmax=2)

f.sp <- RFsimulate(model=RMexp(), x=x, n=n)
f.old <- RFsimulate(model=RMexp(), x=x, n=n, spConform=FALSE)
all.equal(RFspDataFrame2conventional(f.sp)$data, f.old, check.attributes=FALSE) ## TRUE

RandomFields

Simulation and Analysis of Random Fields

v3.3.10

GPL (>= 3)

Authors

Martin Schlather [aut, cre], Alexander Malinowski [aut], Marco Oesting [aut], Daphne Boecker [aut], Kirstin Strokorb [aut], Sebastian Engelke [aut], Johannes Martini [aut], Felix Ballani [aut], Olga Moreva [aut], Jonas Auel[ctr], Peter Menck [ctr], Sebastian Gross [ctr], Ulrike Ober [ctb], Paulo Ribeiro [ctb], Brian D. Ripley [ctb], Richard Singleton [ctb], Ben Pfaff [ctb], R Core Team [ctb]

Initial release