mvPot (version 0.1.7)

simulPareto: Simulate log-Gaussian Pareto random vectors

Description

simulPareto provides n replicates of the multivariate Pareto distribution associated to log-Gaussian random function with semi-variogram vario.

Usage

simulPareto(n, loc, vario, nCores = 1, cl = NULL)

Value

List of n random vectors drawn from a multivariate Pareto distribution with semi-variogram vario.

Arguments

n

Number of replicates desired.

loc

Matrix of coordinates as given by expand.grid().

vario

Semi-variogram function.

nCores

Number of cores used for the computation

cl

Cluster instance as created by makeCluster of the parallel package. Make sure the random number generator has been properly initialized with clusterSetRNGStream().

Details

The algorithm used here is based on the spectral representation of the Brown--Resnick model as described in Dombry et al. (2015). It provides n replicates conditioned that mean(x) > 1 on the unit Frechet scale.

References

Dombry, C., Engelke, S. and M. Oesting. Exact simulation of max-stable processes. Biometrika, 103(2), 303-317.

Examples

#Define variogram function
vario <- function(h){
   1 / 2 * norm(h,type = "2")^1.5
}

#Define locations
loc <- expand.grid(1:4, 1:4)

#Simulate data
obs <- simulPareto(100, loc, vario)