Learn R Programming

RandomFields (version 3.0.5)

RMbr2: Transformation among tail correlation functions

Description

These functions can be used to model different max-stable processes with the same tail correlation functions.

$$C(h) = \cos(\pi (1 - 2\Phi(\sqrt{\gamma(h) / 8}) ) )$$ and $$C(h) =1 - 2 (1 - 2 \Phi(\sqrt{\gamma(h) / 8}) )^2 ,$$ Here, $\Phi$ is the standard normal distribution function, and $\gamma$ is a variogram with sill $8(erf^{-1}(1/sqrt 2))^2 = 1.8197$ and $8(erf^{-1}(1/2))^2 = 4.425098$, respectively.

Usage

RMbr2bg(phi, var, scale, Aniso, proj)
RMbr2eg(phi, var, scale, Aniso, proj)

Arguments

phi
covariance function of class RMmodel.
var,scale,Aniso,proj
optional parameters; same meaning for any RMmodel. If not passed, the above covariance function remains unmodified.

Value

Details

  • RMbr2bg binary random fieldRPbernoullisimulated withRMbr2bg(RMmodel())has a uncentered covariance function that equals
    1. the tail correlation function of the max-stable process constructed with this binary random field
    2. the tail correlation function of Brown-Resnick process with variogramRMmodel.
  • RMbr2eg The extremal Gaussian modelRPschlathersimulated withRMbr2eg(RMmodel())has tail correlation function that equals the tail correlation function of Brown-Resnick process with variogramRMmodel.

References

Strokorb, K. (2012) PhD thesis.

See Also

RPbernoulli, RPbrownresnick, RPschlather, RMmodel.

Examples

Run this code
set.seed(0)
## see 'maxstableAdvanced' for some examples

Run the code above in your browser using DataLab