Optimal copula transport for clustering multivariate time series

Gautier Marti; Frank Nielsen; Philippe Donnat

ICASSP 2016accepted0 citations

Optimal copula transport for clustering multivariate time series

Gautier Marti, Frank Nielsen, Philippe Donnat

Abstract

This paper presents a new methodology for clustering multivariate time series leveraging optimal transport between copulas. Copulas are used to encode both (i) intra-dependence of a multivariate time series, and (ii) inter-dependence between two time series. Then, optimal copula transport allows us to define two distances between multivariate time series: (i) one for measuring intra-dependence dissimilarity, (ii) another one for measuring inter-dependence dissimilarity based on a new multivariate dependence coefficient which is robust to noise, deterministic, and which can target specified dependencies.

BibTeX

@inproceedings{icassp2016_optimalcopulatra,
  title = {Optimal copula transport for clustering multivariate time series},
  author = {Gautier Marti and Frank Nielsen and Philippe Donnat},
  booktitle = {ICASSP 2016},
  year = {2016}
}