AbstractTo capture the extremal behaviour of complex
environmental phenomena in practice, flexible techniques for
modelling tail behaviour are required. In this paper, we
introduce a variety of such methods, which were used by the
Lancopula Utopiversity team to tackle the EVA (2023) Conference
Data Challenge. This data challenge was split into four
challenges, labelled C1-C4. Challenges C1 and C2 comprise
univariate problems, where the goal is to estimate extreme
quantiles for a non-stationary time series exhibiting several
complex features. For these, we propose a flexible modelling
technique, based on generalised additive models, with
diagnostics indicating generally good performance for the
observed data. Challenges C3 and C4 concern multivariate
problems where the focus is on estimating joint probabilities.
For challenge C3, we propose an extension of available models in
the multivariate literature and use this framework to estimate
joint probabilities in the presence of non-stationary
dependence. Finally, for challenge C4, which concerns a
50-dimensional random vector, we employ a clustering technique
to achieve dimension reduction and use a conditional modelling
approach to estimate extremal probabilities across independent
groups of variables.
%0 Journal Article
%1 Andre2024-eb
%A André, L\'ıdia Maria
%A Campbell, Ryan
%A D'Arcy, Eleanor
%A Farrell, Aiden
%A Healy, Dáire
%A Kakampakou, Lydia
%A Murphy, Conor
%A Murphy-Barltrop, Callum John Rowlandson
%A Speers, Matthew
%D 2024
%I Springer Science and Business Media LLC
%J Extremes (Boston)
%K Extreme Utopia Yaff estimating events methods rare value
%T Extreme value methods for estimating rare events in Utopia
%X AbstractTo capture the extremal behaviour of complex
environmental phenomena in practice, flexible techniques for
modelling tail behaviour are required. In this paper, we
introduce a variety of such methods, which were used by the
Lancopula Utopiversity team to tackle the EVA (2023) Conference
Data Challenge. This data challenge was split into four
challenges, labelled C1-C4. Challenges C1 and C2 comprise
univariate problems, where the goal is to estimate extreme
quantiles for a non-stationary time series exhibiting several
complex features. For these, we propose a flexible modelling
technique, based on generalised additive models, with
diagnostics indicating generally good performance for the
observed data. Challenges C3 and C4 concern multivariate
problems where the focus is on estimating joint probabilities.
For challenge C3, we propose an extension of available models in
the multivariate literature and use this framework to estimate
joint probabilities in the presence of non-stationary
dependence. Finally, for challenge C4, which concerns a
50-dimensional random vector, we employ a clustering technique
to achieve dimension reduction and use a conditional modelling
approach to estimate extremal probabilities across independent
groups of variables.
@article{Andre2024-eb,
abstract = {AbstractTo capture the extremal behaviour of complex
environmental phenomena in practice, flexible techniques for
modelling tail behaviour are required. In this paper, we
introduce a variety of such methods, which were used by the
Lancopula Utopiversity team to tackle the EVA (2023) Conference
Data Challenge. This data challenge was split into four
challenges, labelled C1-C4. Challenges C1 and C2 comprise
univariate problems, where the goal is to estimate extreme
quantiles for a non-stationary time series exhibiting several
complex features. For these, we propose a flexible modelling
technique, based on generalised additive models, with
diagnostics indicating generally good performance for the
observed data. Challenges C3 and C4 concern multivariate
problems where the focus is on estimating joint probabilities.
For challenge C3, we propose an extension of available models in
the multivariate literature and use this framework to estimate
joint probabilities in the presence of non-stationary
dependence. Finally, for challenge C4, which concerns a
50-dimensional random vector, we employ a clustering technique
to achieve dimension reduction and use a conditional modelling
approach to estimate extremal probabilities across independent
groups of variables.},
added-at = {2025-01-07T14:37:35.000+0100},
author = {Andr{\'e}, L{\'\i}dia Maria and Campbell, Ryan and D'Arcy, Eleanor and Farrell, Aiden and Healy, D{\'a}ire and Kakampakou, Lydia and Murphy, Conor and Murphy-Barltrop, Callum John Rowlandson and Speers, Matthew},
biburl = {https://puma.scadsai.uni-leipzig.de/bibtex/2a53866754687ff07286a9b7c1da54d7f/scadsfct},
copyright = {https://creativecommons.org/licenses/by/4.0},
interhash = {1133a6b787137e61f58b8c88965089c6},
intrahash = {a53866754687ff07286a9b7c1da54d7f},
journal = {Extremes (Boston)},
keywords = {Extreme Utopia Yaff estimating events methods rare value},
language = {en},
month = nov,
publisher = {Springer Science and Business Media LLC},
timestamp = {2025-01-31T11:52:11.000+0100},
title = {Extreme value methods for estimating rare events in Utopia},
year = 2024
}