Your copula is a classifier in disguise: classification-based copula density estimation

TL;DR

This paper introduces a novel approach to copula density estimation by framing it as a classification problem, enabling effective estimation and sampling of complex dependencies in data.

Contribution

It reinterprets copula density estimation as a discriminative classification task, connecting it with density ratio estimation and providing theoretical guarantees.

Findings

Outperforms existing copula estimators in density evaluation

Effective in high-dimensional datasets

Achieves theoretical guarantees similar to maximum likelihood

Abstract

We propose reinterpreting copula density estimation as a discriminative task. Under this novel estimation scheme, we train a classifier to distinguish samples from the joint density from those of the product of independent marginals, recovering the copula density in the process. We derive equivalences between well-known copula classes and classification problems naturally arising in our interpretation. Furthermore, we show our estimator achieves theoretical guarantees akin to maximum likelihood estimation. By identifying a connection with density ratio estimation, we benefit from the rich literature and models available for such problems. Empirically, we demonstrate the applicability of our approach by estimating copulas of real and high-dimensional datasets, outperforming competing copula estimators in density evaluation as well as sampling.