Tree-based conditional copula estimation

This article proposes a regression tree procedure to estimate conditional copulas. The associated algorithm determines classes of observations based on covariate values and fits a simple parametric copula model on each class. The association parameter changes from one class to another, allowing for...

Full description

Saved in:
Bibliographic Details
Main Authors: Bonacina Francesco, Lopez Olivier, Thomas Maud
Format: Article
Language:English
Published: De Gruyter 2025-02-01
Series:Dependence Modeling
Subjects:
Online Access:https://doi.org/10.1515/demo-2024-0010
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850198003319570432
author Bonacina Francesco
Lopez Olivier
Thomas Maud
author_facet Bonacina Francesco
Lopez Olivier
Thomas Maud
author_sort Bonacina Francesco
collection DOAJ
description This article proposes a regression tree procedure to estimate conditional copulas. The associated algorithm determines classes of observations based on covariate values and fits a simple parametric copula model on each class. The association parameter changes from one class to another, allowing for non-linearity in the dependence structure modeling. It also allows the definition of classes of observations on which the so-called “simplifying assumption” holds reasonably well. When considering observations belonging to a given class separately, the association parameter no longer depends on the covariates according to our model. In this article, we derive asymptotic consistency results for the regression tree procedure and show that the proposed pruning methodology, i.e., the model selection techniques selecting the appropriate number of classes, is optimal in some sense. Simulations provide finite sample results, and an analysis of data of cases of human influenza presents the practical behavior of the procedure.
format Article
id doaj-art-be5a494b095348aebc28529a13da4b8b
institution OA Journals
issn 2300-2298
language English
publishDate 2025-02-01
publisher De Gruyter
record_format Article
series Dependence Modeling
spelling doaj-art-be5a494b095348aebc28529a13da4b8b2025-08-20T02:12:59ZengDe GruyterDependence Modeling2300-22982025-02-01131437310.1515/demo-2024-0010Tree-based conditional copula estimationBonacina Francesco0Lopez Olivier1Thomas Maud2Sorbonne Université, CNRS, Laboratoire de Probabilités, Statistique et Modélisation, LPSM, 4 place Jussieu, F-75005 Paris, FranceCREST Laboratory, CNRS, Groupe des Écoles Nationales d’Économie et Statistique, Ecole Polytechnique, Institut Polytechnique de Paris, 5 avenue Henry Le Chatelier91120 Palaiseau, FranceSorbonne Université, CNRS, Laboratoire de Probabilités, Statistique et Modélisation, LPSM, 4 place Jussieu, F-75005 Paris, FranceThis article proposes a regression tree procedure to estimate conditional copulas. The associated algorithm determines classes of observations based on covariate values and fits a simple parametric copula model on each class. The association parameter changes from one class to another, allowing for non-linearity in the dependence structure modeling. It also allows the definition of classes of observations on which the so-called “simplifying assumption” holds reasonably well. When considering observations belonging to a given class separately, the association parameter no longer depends on the covariates according to our model. In this article, we derive asymptotic consistency results for the regression tree procedure and show that the proposed pruning methodology, i.e., the model selection techniques selecting the appropriate number of classes, is optimal in some sense. Simulations provide finite sample results, and an analysis of data of cases of human influenza presents the practical behavior of the procedure.https://doi.org/10.1515/demo-2024-0010conditional copularegression treesasymptotic theory62g0862g2062h3062p10
spellingShingle Bonacina Francesco
Lopez Olivier
Thomas Maud
Tree-based conditional copula estimation
Dependence Modeling
conditional copula
regression trees
asymptotic theory
62g08
62g20
62h30
62p10
title Tree-based conditional copula estimation
title_full Tree-based conditional copula estimation
title_fullStr Tree-based conditional copula estimation
title_full_unstemmed Tree-based conditional copula estimation
title_short Tree-based conditional copula estimation
title_sort tree based conditional copula estimation
topic conditional copula
regression trees
asymptotic theory
62g08
62g20
62h30
62p10
url https://doi.org/10.1515/demo-2024-0010
work_keys_str_mv AT bonacinafrancesco treebasedconditionalcopulaestimation
AT lopezolivier treebasedconditionalcopulaestimation
AT thomasmaud treebasedconditionalcopulaestimation