On logarithmically optimal exact simulation of max-stable and related random fields on a compact set

Zhipeng Liu, Jose H. Blanchet, A.b. Dieker, Thomas Mikosch

2 Citations (Scopus)
15 Downloads (Pure)

Abstract

We consider the random field M(t) = nsup ≥1 {− log An + Xn(t)}, t ∈ T, for a set T ⊂ Rm, where (Xn) is an i.i.d. sequence of centered Gaussian random fields on T and 0 < A1 < A2 < · · · are the arrivals of a general renewal process on (0, ∞), independent of (Xn). In particular, a large class of max-stable random fields with Gumbel marginals have such a representation. Assume that one needs c(d) = c({t1, . . ., td}) function evaluations to sample Xn at d locations t1, . . ., td ∈ T . We provide an algorithm which samples M(t1), . . ., M(td) with complexity O(c(d)1+o(1)) as measured in the Lp norm sense for any p ≥ 1. Moreover, if Xn has an a.s. converging series representation, then M can be a.s. approximated with error δ uniformly over T and with complexity O(1/(δ log(1/δ))1/α), where α relates to the Hölder continuity exponent of the process Xn (so, if Xn is Brownian motion, α = 1/2).

Original languageEnglish
JournalBernoulli
Volume25
Issue number4A
Pages (from-to)2949-2981
ISSN1350-7265
DOIs
Publication statusPublished - 2019

Fingerprint

Dive into the research topics of 'On logarithmically optimal exact simulation of max-stable and related random fields on a compact set'. Together they form a unique fingerprint.

Cite this