RFspatialPointsDataFrame-class: Class "RFspatialPointsDataFrame"

In RandomFields: Simulation and Analysis of Random Fields

Description

Class for spatial attributes that have spatial or spatio-temporal locations (at least of dimension 2) that are not on a grid. Direct extension of class SpatialPointsDataFrame from the sp-package. See sp2RF for an explicit transformation.

Usage

## S4 method for signature 'RFspatialPointsDataFrame'
RFspDataFrame2conventional(obj) 

Arguments

obj	an RFspatialPointsDataFrame object

Creating Objects

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

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

coords.nrs:

See SpatialPointsDataFrame.

coords:

matrix of coordinates (each row is a point); in case of SpatialPointsDataFrame an object of class SpatialPoints is also allowed, see SpatialPoints.

bbox:

matrix specifying the bounding box

proj4string:

object of class CRS; projection

Extends

Class SpatialPointsDataFrame, directly. Class SpatialPoints, by class SpatialPointsDataFrame. Class Spatial, by class SpatialPoints.

Methods

plot

signature(obj = "RFspatialPointsDataFrame"): generates nice 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("RFspatialPointsDataFrame")

show

signature(x = "RFspatialPointsDataFrame"): uses the show-method for class SpatialPointsDataFrame

print

signature(x = "RFspatialPointsDataFrame"): identical to show-method

RFspDataFrame2conventional

signature(obj = "RFspatialPointsDataFrame"): 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)]

coordinates

signature(x = "RFspatialPointsDataFrame"): returns the coordinates

[

signature(x = "RFspatialPointsDataFrame"): selects columns of data-slot; returns an object of class RFspatialPointsDataFrame

[<-

signature(x = "RFspatialPointsDataFrame"): replaces columns of data-slot; returns an object of class RFspatialPointsDataFrame

as

signature(x = "RFspatialPointsDataFrame"): converts into other formats, only implemented for target class RFspatialGridDataFrame

cbind

signature(...): if arguments have identical topology, combine their attribute values

range

signature(x = "RFspatialPointsDataFrame"): returns the range

hist

signature(x = "RFspatialPointsDataFrame"): plots histogram

as.matrix

signature(x = "RFspatialPointsDataFrame"): converts data-slot to matrix

as.array

signature(x = "RFspatialPointsDataFrame"): converts data-slot to array

as.vector

signature(x = "RFspatialPointsDataFrame"): converts data-slot to vector

as.data.frame

signature(x = "RFspatialPointsDataFrame"): 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

See Also

RFspatialGridDataFrame-class, which is for point locations that are on a grid, RFpointsDataFrame-class which is for one-dimensional locations, RFsp, sp2RF

Examples

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

x <- cbind(runif(50), runif(50))
f <- RFsimulate(model=RMexp(), x=x, n=3)

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

plot(f, nmax=2)

f2 <- RFsimulate(model=RMwhittle(nu=1.2, scale=10), x=cbind(x,x), n=4)
plot(f2, MARGIN=c(3,4), nmax=2)

f.sp <- RFsimulate(model=RMexp(), x=x, n=3)
f.old <- RFsimulate(model=RMexp(), x=x, n=3, spConform=FALSE)
all.equal(RFspDataFrame2conventional(f.sp)$data, f.old, check.attributes=FALSE) ## TRUE

RandomFields documentation built on Jan. 19, 2022, 1:06 a.m.