A mixture model for multivariate extremes

The spectral density function plays a key role in ﬁtting the tail of multivariate extremal data and so in estimating probabilities of rare events. This function satisﬁes moment constraints but unlike the univariate extreme value distributions has no simple parametric form. Parameterized subfamilies of spectral densities have been suggested for use in applications, and nonparametric estimation procedures have been proposed, but semiparametric models for multivariate extremes have hitherto received little attention. We show that mixtures of Dirichlet distributions satisfying the moment constraints are weakly dense in the class of all nonparametric spectral densities, and discuss frequentist and Bayesian inference in this class based on the EM algorithm and reversible jump Markov chain Monte Carlo simulation. We illustrate the ideas using simulated and real data.

Mixture models for multivariate extremes

Marc-Olivier Boldi

This thesis is a contribution to multivariate extreme value statistics. The tail of a multivariate distribution function is characterized by its spectral distribution, for which we propose a new semi-parametric model based on mixtures of Dirichlet distributions. To estimate the components of this model, reversible jump Monte Carlo Markov chain and EM algorithms are developed. Their performances are illustrated on real and simulated data, obtained using new representations of the extremal logistic and Dirichlet models. In parallel with the estimation of the spectral distribution, extreme value statistic machinery requires the selection of a threshold in order to classify data as extreme or not. This selection is achieved by a new method based on heuristic arguments. It allows a selection independent of the dimension of the data. Its performance is illustrated on real and simulated data. Primal scientific interests behind a multivariate extreme value analysis reside in the estimation of quantiles of rare events and in the exploration of the dependence structure, for which the estimation of the spectral measure is a means rather than an end. These two issues are addressed. For the first, a Monte Carlo method is developed based on simulation of extremes. It is compared with classical and new methods of the literature. For the second one, an original conditional dependence analysis is proposed, which enlightens various aspects of the dependence structure of the data. Examples using real data sets are given. In the last part, the semi-parametric model and the presented methods are extended to spatial extremes. It is made possible by considering the spectral distribution as the distribution of a random probability, an original viewpoint adopted throughout this thesis. Classical multivariate extremes are extended to extremes of random measures. The application is illustrated on rainfall data in China.

EPFL2004

Inference on the Angular Distribution of Extremes

Claudio Andri Semadeni

The spectral distribution plays a key role in the statistical modelling of multivariate extremes, as it defines the dependence structure of multivariate extreme-value distributions and characterizes the limiting distribution of the relative sizes of the components of large multivariate observations. No parametric family captures all possible types of multivariate dependence, and numerous parametric models have been proposed. Inference on the spectral distribution is typically based on the pseudo-angles of

large' observations under the assumption that they follow the spectral distribution. There has been little attention on studying the impact of this approximation on inference, and it turns out that it can yield significantly biased estimates. We provide a characterization of the angular distribution of excesses corresponding to the distribution of pseudo-angles of

large'observations that improves direct inference on the spectral distribution in the bivariate setting. Extremal dependence is at the heart of extreme value modelling and numerous measures to quantify it have been proposed in the literature. In many applications, datasets seem to exhibit asymmetry in the dependence between the variables. Many parametric multivariate extreme-value models can accommodate asymmetry in the sense that the spectral density can be asymmetric, resulting in a non-exchangeable dependence structure. There has been little attention paid to quantifying asymmetry at extreme levels, which can be useful for diagnosis and model checking. We propose a coefficient of extremal asymmetry that quantifies the asymmetry at extreme levels for pairs of variables. We also propose two non-parametric estimators of the coefficient of extremal asymmetry and compare their properties through numerical simulation. The two estimators have diametrically opposed bias-variance trade-offs. The estimator based on maximum empirical likelihood performs well and is nearly unbiased.

EPFL2020

Fast high-dimensional Bayesian classification and clustering

Vahid Partovi Nia

We introduce a fast approach to classification and clustering applicable to high-dimensional continuous data, based on Bayesian mixture models for which explicit computations are available. This permits us to treat classification and clustering in a single framework, and allows calculation of unobserved class probability. The new classifier is robust to adding noise variables as a drawback of the built-in spike-and-slab structure of the proposed Bayesian model. The usefulness of classification using our method is shown on metabololomic example, and on the Iris data with and without noise variables. Agglomerative hierarchical clustering is used to construct a dendrogram based on the posterior probabilities of particular partitions, to provide a dendrogram with a probabilistic interpretation. An extension to variable selection is proposed which summarises the importance of variables for classification or clustering and has probabilistic interpretation. Having a simple model provides estimation of the model parameters using maximum likelihood and therefore yields a fully automatic algorithm. The new clustering method is applied to metabolomic, microarray, and image data and is studied using simulated data motivated by real datasets. The computational difficulties of the new approach are discussed, solutions for algorithm acceleration are proposed, and the written computer code is briefly analysed. Simulations shows that the quality of the estimated model parameters depends on the parametric distribution assumed for effects, but after fixing the model parameters to reasonable values, the distribution of the effects influences clustering very little. Simulations confirms that the clustering algorithm and the proposed variable selection method is reliable when the model assumptions are wrong. The new approach is compared with the popular Bayesian clustering alternative, MCLUST, fitted on the principal components using two loss functions in which our proposed approach is found to be more efficient in almost every situation.

EPFL2009