Variants of the graph dependent model in extreme value theory
The author considers the general graph dependent model introduced by him in an earlier paper [Ann. Math. Statist. 43, 516-521 (1972; Zbl 0238.62060)]. It covers several known dependent extreme value models (e.g. m-dependence, several kinds of mixing concepts). With a set \(X_ 1,X_ 2,...,X_ n\) of n random variables, a graph is associated whose vertices are the integers 1,2,...,n, and two integers i and j are connected with an edge if the events \(\{X_ i>x\}\) and \(\{X_ j>x\}\) do not become ``almost independent for ``large x. In this paper the limiting distribution of \(Z_ n=\max \{X_ 1,...,X_ n\}\) is determined under appropriate conditions on the edge set \(E_ n\) of the graph. By specializing \(E_ n\), one gets a variate limiting distribution in one theorem. The results refine those of the author [op. cit.] on this kind of dependence.
- A note on the functional law of iterated logarithm for maxima of Gaussian sequences
- Extreme Values in Samples from m-Dependent Stationary Stochastic Processes
- Extreme Values in Uniformly Mixing Stationary Stochastic Processes
- Limit Theorems for the Maximum Term in Stationary Sequences
- On the Distribution of the Maximum of Random Variables
- New exploratory tools for extremal dependence: \(\chi \) networks and annual extremal networks
- On the relationship between the discrete and continuous bounding moment problems and their numerical solutions
- One-component regular variation and graphical modeling of extremes
- scientific article; zbMATH DE number 850345 (Why is no real title available?)
- Graphical Models for Extremes
- Extremal independence in discrete random systems
This page was built for publication: Variants of the graph dependent model in extreme value theory
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3805536)