fdWasserstein

CRANfdWassersteinMaRDI QIDQ5976290FDOQ5976290

Application of Optimal Transport to Functional Data Analysis

Guido Masarotto, Valentina Masarotto

Last update: 6 February 2024

Copyright license: GNU General Public License, version 3.0

Software version identifier: 1.0

Description

These functions were developed to support statistical analysis on functional covariance operators. The package contains functions to: - compute 2-Wasserstein distances between Gaussian Processes as in Masarotto, Panaretos & Zemel (2019) <doi:10.1007/s13171-018-0130-1>; - compute the Wasserstein barycenter (Frechet mean) as in Masarotto, Panaretos & Zemel (2019) <doi:10.1007/s13171-018-0130-1>; - perform analysis of variance testing procedures for functional covariances and tangent space principal component analysis of covariance operators as in Masarotto, Panaretos & Zemel (2022) <arXiv:2212.04797>. - perform a soft-clustering based on the Wasserstein distance where functional data are classified based on their covariance structure as in Masarotto & Masarotto (2023) <doi:10.1111/sjos.12692>.

Cites work

• Procrustes metrics on covariance operators and optimal transportation of Gaussian processes
• Covariance‐based soft clustering of functional data based on the Wasserstein–Procrustes metric
• Transportation-Based Functional ANOVA and PCA for Covariance Operators

This page was built for software: fdWasserstein