• Corpus ID: 228375991

Structure learning for extremal tree models

  title={Structure learning for extremal tree models},
  author={Sebastian Engelke and Stanislav Volgushev},
  journal={arXiv: Methodology},
Extremal graphical models are sparse statistical models for multivariate extreme events. The underlying graph encodes conditional independencies and enables a visual interpretation of the complex extremal dependence structure. For the important case of tree models, we develop a data-driven methodology for learning the graphical structure. We show that sample versions of the extremal correlation and a new summary statistic, which we call the extremal variogram, can be used as weights for a… 
Total positivity in multivariate extremes
An estimator for the parameters of the Hüsler–Reiss distribution under EMTP2 is proposed as the solution of a convex optimization problem with Laplacian constraint and it is proved that this estimator is consistent and typically yields a sparse model with possibly non-decomposable extremal graphical structure.
Extreme event propagation using counterfactual theory and vine copulas
Understanding multivariate extreme events play a crucial role in managing the risks of complex systems since extremes are governed by their own mechanisms. Conditional on a given variable exceeding a


Structure Learning in Graphical Modeling
This work discusses methods such as the graphical lasso and neighborhood selection for undirected graphical models, and the PC algorithm and score-based search methods for directed graphical models (or Bayesian networks), and reviews extensions that account for effects of latent variables and heterogeneous data sources.
Inference on extremal dependence in a latent Markov tree model attracted to a H\"usler-Reiss distribution
The structured Husler-Reiss distribution is found to fit the observed extremal dependence well, and the fitted model confirms the importance of flow-connectedness for the strength of dependence between high water levels, even for locations at large distance apart.
Concentration bounds for the extremal variogram
In extreme value theory, the extremal variogram is a summary of the tail dependence of a random vector. It is a central ingredient for learning extremal tree structures (Engelke and Volgushev, 2020)
Sparse Nonparametric Graphical Models
Two approaches to building more flexible graphical models are discussed, one allows arbitrary graphs and a nonparametric extension of the Gaussian; the other uses kernel density estimation and restricts the graphs to trees and forests.
Dependence Measures for Extreme Value Analyses
Quantifying dependence is a central theme in probabilistic and statistical methods for multivariate extreme values. Two situations are possible: one where, in a limiting sense, the extremes are
Extremal behaviour of aggregated data with an application to downscaling
&NA; The distribution of spatially aggregated data from a stochastic process X may exhibit tail behaviour different from that of its marginal distributions. For a large class of aggregating
Bayesian Networks for Max-Linear Models
It is argued that the structure of a minimal network asymptotically can be identified completely from observational data.
Multivariate peaks over thresholds models
Multivariate peaks over thresholds modelling based on generalized Pareto distributions has up to now only been used in few and mostly two-dimensional situations. This paper contributes theoretical
Extremal dependence measure and extremogram: the regularly varying case
The dependence of large values in a stochastic process is an important topic in risk, insurance and finance. The idea of risk contagion is based on the idea of large value dependence. The Gaussian
An M-Estimator for Tail Dependence in Arbitrary Dimensions
Consider a random sample in the max-domain of attraction of a multivariate extreme value distribution such that the dependence structure of the attractor belongs to a parametric model. A new