Learn R Programming

HiDimMaxStable (version 0.1.1)

Inference on High Dimensional Max-Stable Distributions

Description

Inference of high dimensional max-stable distributions, from the paper "Likelihood based inference for high-dimensional extreme value distributions", by A. Bienvenüe and C. Robert, arXiv:1403.0065 [stat.AP].

Copy Link

Version

Install

install.packages('HiDimMaxStable')

Monthly Downloads

10

Version

0.1.1

License

GPL (>= 2)

Maintainer

Alexis Bienvene

Last Published

January 14th, 2015

Functions in HiDimMaxStable (0.1.1)

excess.censor

Transforms data to normalized exceedances with censoring
spatial

Spatial models
simultoccur.l

Likelihood for vectors of componentwise maxima with additional information on maxima occurences
rSchlatherExcess

Simulation of vectors in the maximum domain of attraction of a spatial Schlather max-stable distribution
excess.l

Likelihood for vectors of exceedance with censored components
maxlik

Maximum likelihood estimation
maxblocks

Computes the normalized componentwise maxima with their occurrences for several blocks
margin

Margin distributions
build.clusters.spatial

Builds clusters with a given maximum size using a k-means clustering.
maxstable.l.clusters

Partition-composite likelihood for multivariate max-stable distributions
spatial-class

spatial class
select.mean

Selects vectors for which the mean of the components is larger than a threshold
plot3d.densgrid

3D visualisation of the computed values of the likelihood function on a grid.
mubz.copula

$mu(B,z)$ for the copula model
mubz.lnormal

$mu(B,z)$ for the Log-normal model
rCMS

Simulation of vectors in the maximum domain of attraction of an homogeneous clustered max-stable distribution
mubz.normal

$mu(B,z)$ for the Gaussian model
maxgrid

Identifies the coordinates of the maximum on a grid
dens.grid

Computes the likelihood function on a grid of parameters
margin-class

margin class