Dynamic Networks with Multi-scale Temporal Structure

TL;DR

This paper introduces a multi-scale dynamic network modeling approach for non-stationary multivariate time series, capturing local temporal structures and interactions with sparsity, demonstrated on neuroscience MEG data.

Contribution

It combines recursive dyadic partitioning with penalized likelihood for dynamic neighborhood selection, offering a novel framework for non-stationary time series analysis.

Findings

Effective detection of local network changes at multiple temporal resolutions

Sparsity-promoting penalty improves interpretability of dynamic networks

Successful application to neuroscience MEG data

Abstract

We describe a novel method for modeling non-stationary multivariate time series, with time-varying conditional dependencies represented through dynamic networks. Our proposed approach combines traditional multi-scale modeling and network based neighborhood selection, aiming at capturing temporally local structure in the data while maintaining sparsity of the potential interactions. Our multi-scale framework is based on recursive dyadic partitioning, which recursively partitions the temporal axis into finer intervals and allows us to detect local network structural changes at varying temporal resolutions. The dynamic neighborhood selection is achieved through penalized likelihood estimation, where the penalty seeks to limit the number of neighbors used to model the data. We present theoretical and numerical results describing the performance of our method, which is motivated and illustrated using task-based magnetoencephalography (MEG) data in neuroscience.