Trasgu
A wrapper for simplified vine copula fitting with Chimera across local and high-performance computing platforms
Valvanuz Fernández-Quiruelas (Universidad de Cantabria)
Patricia Mares-Nasarre (TU Delft - Civil Engineering & Geosciences)
Cornelis Marcel Pieter ’t Hart (TU Delft - Civil Engineering & Geosciences, Haskoning)
Oswaldo Morales-Nápoles (TU Delft - Civil Engineering & Geosciences)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Vine copulas represent multivariate probability distributions through sequences of trees whose edges are modelled with bivariate copulas. Chimera is a catalogue developed at TU Delft that contains all possible regular-vine tree sequences with four to eight nodes, represented by 663,206,904 matrices. Exhaustively fitting these matrices provides a reference for evaluating structure-selection algorithms based on heuristics, but requires managing millions of independent fits. Trasgu is an open-source Python package developed to make such studies reproducible on personal computers and high-performance computing systems. We illustrate the use of Trasgu through a simulation study and a real-world application with field data on ship-induced waves.