RandomFields (version 3.1.12)

RMbernoulli: 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.

Value

• RMbernoulli returns an object of class RMmodel.

Details

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

References

Ballani, Schlather

See Also

RPbernoulli RMmodel, RFsimulate,

Examples

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

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")



FinalizeExample()