RandomFields (version 3.1.12)

RMschlather: Covariance Model for binary field based on Gaussian field

Description

RMschlather gives the tail correlation function of the extremal Gaussian process, i.e.

$$C(h) = 1 - \sqrt{ (1-\phi(h)/\phi(0)) / 2 }$$

where $\phi$ is the covariance of a stationary Gaussian field.

Usage

RMschlather(phi, var, scale, Aniso, proj)

Arguments

phi

covariance function of class RMmodel.

var,scale,Aniso,proj

optional arguments; same meaning for any RMmodel. If not passed, the above covariance function remains unmodified.

Value

• RMschlather returns an object of class RMmodel

Details

This model yields the tail correlation function of the field that is returned by RPschlather

See Also

RPschlather 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()
## This examples considers an extremal Gaussian random field
## with Gneiting's correlation function.

## first consider the covriance model and its corresponding tail
## corrlation function
model <- RMgneiting()
plot(model, model.tail.corr.fct=RMschlather(model),  xlim=c(0, 5))


## the extremal Gaussian field with the above underlying
## correlation function that has the above tail correlation function tcf
x <- seq(0, 10, 0.1)
z <- RFsimulate(RPschlather(model), x)
plot(z)

## Note that in RFsimulate R-P-schlather was called, not R-M-schlather.
## The following lines give a Gaussian random field with corrlation
## function equal to the above tail correlation function.
z <- RFsimulate(RMschlather(model), x)
plot(z)


FinalizeExample()