Exact simulation from max-stable processes and multivariate extreme value distributions for various parametric models.
The package allows exact generation of multivariate extreme value vectors or max-stable processes. For the latter, the user can provide a variogram function along with a set of locations that serve as input. Models implemented include the 1-parameter logistic and negative logistic as described in the article, the bilogistic and Coles and Tawn extremal Dirichlet model using the algorithm of Boldi (2009) and the Dirichlet mixture. The extremal Student and Husler-Reiss (Brown-Resnick) models are also implemented.
Other features of the package include threshold diagnostic tests, empirical likelihood estimation for the spectral measure,
bias-correction and tangent exponential model approximations for the GP and GEV distributions, extended generalized Pareto distributions, information matrix test and penultimate approximations. Many of these are not
found elsewhere. mev complements packages evd and ismev.
Dombry, Engelke and Oesting (2016). Exact simulation of max-stable processes, Biometrika, 103(2), 303--317.
Boldi (2009). A note on the representation of parametric models for multivariate extremes. Extremes 12, 211--218.