R: Covariance Model for binary field based on a Gaussian field

RMbernoulli {RandomFields}

R Documentation

Covariance Model for binary field based on a Gaussian field

Description

RMbernoulli gives the centered correlation function of a binary field, obtained by thresholding a Gaussian field.

Usage

RMbernoulli(phi, threshold, correlation, centred, var, scale, Aniso, proj)

Arguments

`phi`	covariance function of class `RMmodel`.
`threshold`	real valued threshold, see `RPbernoulli`. Currently, only `threshold=0.0` is possible. Default: 0.
`correlation`	logical. If `FALSE` the corresponding covariance function is returned. Default: `TRUE`.
`centred`	logical. If `FALSE` the uncentred covariance is returned. Default: `TRUE`.
`var,scale,Aniso,proj`	optional arguments; same meaning for any `RMmodel`. If not passed, the above covariance function remains unmodified.

Details

This model yields the covariance function of the field that is returned by RPbernoulli.

Value

RMbernoulli returns an object of class RMmodel.

Note

Previous to version 3.0.33 the covariance function was returned, not the correlation function.

Author(s)

Martin Schlather, schlather@math.uni-mannheim.de, http://ms.math.uni-mannheim.de

References

Ballani, Schlather

Examples

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

threshold <- 0
x <- seq(0, 5, 0.02)
GaussModel <- RMgneiting()

n <- 1000
z <- RFsimulate(RPbernoulli(GaussModel, threshold=threshold), x=x, n=n)
plot(z)

model <- RMbernoulli(RMgauss(), threshold=threshold, correlation=FALSE)
plot(model, xlim=c(0,5))
z1 <- as.matrix(z)
estim.cov <- apply(z1, 1, function(x) cov(x, z1[1,]))
points(coordinates(z), estim.cov, col="red")

[Package RandomFields version 3.3.4 Index]