UNIGE document Doctoral Thesis
previous document  unige:106918  next document
add to browser collection

Statistical modelling and inference for covariate-dependent extremal dependence

Chavez-Demoulin, Valérie
Defense Thèse de doctorat : Univ. Genève, 2018 - GSEM 57 - 2018/06/28
Abstract In this thesis, we develop new models for covariate-varying tail dependence structures and propose novel techniques for fitting these models to both block maxima and threshold exceedances data, under the assumptions of asymptotic dependence and asymptotic independence. Our proposals for the flexible incorporation of covariate influence rely on the (vector) generalized additive modelling infrastructure, and are established in a parametric setting and a non-parametric setting where we develop projection techniques enabling the reduction of the problem of characterizing joint tail dependences to the modelling of univariate random variables. Inference is performed by penalized maximum likelihood estimation combined, when applicable, with censored likelihood techniques. The performance of the resulting estimators is assessed either through simulation studies or based on asymptotic distributions. The developed methodologies are illustrated on environmental datasets where dependence between large events is linked to a set of covariates describing time as well as characteristics of the measurement sites.
Keywords Angular densityAsymptotic dependenceAsymptotic independenceCovariate-adjustmentExtreme value theoryGeneralized additive modelsMax-stable random vectorsMultivariate extreme valuesPenalized log-likelihoodThreshold exceedancesVector generalized additive models.
URN: urn:nbn:ch:unige-1069182
Full text
Thesis (15.9 MB) - public document Free access
(ISO format)
MHALLA, Linda. Statistical modelling and inference for covariate-dependent extremal dependence. Université de Genève. Thèse, 2018. doi: 10.13097/archive-ouverte/unige:106918 https://archive-ouverte.unige.ch/unige:106918

273 hits



Deposited on : 2018-08-08

Export document
Format :
Citation style :