Dependence structure estimation using Copula Recursive Trees

JOURNAL OF MULTIVARIATE ANALYSIS

卷:185

Article

Laverny, Oskar¹  Masiello, Esterina¹  Maume-Deschamps, Veronique¹  Rulliere, Didier² 

[1] Univ Claude Bernard Lyon 1, Univ Lyon, CNRS UMR 5208, Inst Camille Jordan, F-69622 Villeurbanne, France

[2] Univ Clermont Auvergne, Mines St Etienne, CNRS, UMR LIMOS 6158,Inst Henri Fayol, F-42023 St Etienne, France

关键词: Bagging;    CORT;    Density estimation trees;    Nonparametric estimation;    Patchwork copula;    Piecewise linear copula;    Quadratic program;   

DOI  :  10.1016/j.jmva.2021.104776

来源: Elsevier

【 摘 要 】

We construct the COpula Recursive Tree (CORT) estimator: a flexible, consistent, piecewise linear estimator of a copula, leveraging the patchwork copula formalization and various piecewise constant density estimators. While the patchwork structure imposes a grid, the CORT estimator is data-driven and constructs the (possibly irregular) grid recursively from the data, minimizing a chosen distance on the copula space. The addition of the copula constraints makes usual density estimators unusable, whereas the CORT estimator is only concerned with dependence and guarantees the uniformity of margins. Refinements such as localized dimension reduction and bagging are developed, analyzed, and tested through simulated data. (C) 2021 Elsevier Inc. All rights reserved.

【 授权许可】

Free