Abstract:Time series data sets often contain heterogeneous signals, composed of both continuously changing quantities and discretely occurring events. The coupling between these measurements may provide insights into key underlying mechanisms of the systems under study. To better extract this information, we investigate the asymptotic statistical properties of coupling measures between continuous signals and point processes. We first introduce martingale stochastic integration theory as a mathematical model for a family of statistical quantities that include the phase locking value, a classical coupling measure to characterize complex dynamics. Based on the martingale central limit theorem, we can then derive the asymptotic gaussian distribution of estimates of such coupling measure that can be exploited for statistical testing. Second, based on multivariate extensions of this result and random matrix theory, we establish a principled way to analyze the low-rank coupling between a large number of point processes and continuous signals. For a null hypothesis of no coupling, we establish sufficient conditions for the empirical distribution of squared singular values of the matrix to converge, as the number of measured signals increases, to the well-known Marchenko-Pastur (MP) law, and the largest squared singular value converges to the upper end of the MP support. This justifies a simple thresholding approach to assess the significance of multivariate coupling. Finally, we illustrate with simulations the relevance of our univariate and multivariate results in the context of neural time series, addressing how to reliably quantify the interplay between multichannel local field potential signals and the spiking activity of a large population of neurons.

An invitation to coupling and copulas: with applications to multisensory modeling

From Univariate to Multivariate Coupling Between Continuous Signals and Point Processes: A Mathematical Framework

Coupling functions: Universal insights into dynamical interaction mechanisms

Implicit Copulas: An Overview

The empirical copula process in high dimensions: Stute's representation and applications

Coupling Learning of Complex Interactions

New copula families and mixing properties

Stochastic Monotone Wing Property: A New Dependence Structure for Copulas

Copula modeling for discrete random vectors

Copula-Based Regression with Mixed Covariates

Multiple risk factor dependence structures: Copulas and related properties

Copula-type Estimators for Flexible Multivariate Density Modeling using Mixtures

MULTIVARIATE COMPOSITE COPULAS

Bivariate Copula Decomposition in Terms of Comonotonicity, Countermonotonicity and Independence

Couplings of the Random-Walk Metropolis algorithm

Modeling Stochastic Data Using Copulas For Application in Validation of Autonomous Driving

Time and Space Varying Copulas

Additive Models for Conditional Copulas

The Coupled Rejection Sampler

A copula approach for dependence modeling in multivariate nonparametric time series

Bayesian Inference for Regression Copulas