Inferring Dynamic Hidden Graph Structure in Heterogeneous Correlated Time Series

TL;DR

This paper introduces a windowed variance-correlation metric (WVC) to infer hidden dynamic graph structures from heterogeneous time series, capturing time-varying relationships that other methods overlook.

Contribution

The paper proposes a novel WVC metric for directly inferring dynamic hidden graph structures in heterogeneous time series data.

Findings

WVC effectively captures correlations missed by other methods.

The method can recover hidden relationships in specified time intervals.

It expands the ability to model dynamic graphs with diverse signals.

Abstract

Modeling heterogeneous correlated time series requires the ability to learn hidden dynamic relationships between component time series with possibly varying periodicities and generative processes. To address this challenge, we formulate and evaluate a windowed variance-correlation metric (WVC) designed to quantify time-varying correlations between signals. This method directly recovers hidden relationships in an specified time interval as a weighted adjacency matrix, consequently inferring hidden dynamic graph structure. On simulated data, our method captures correlations that other methods miss. The proposed method expands the ability to learn dynamic graph structure between significantly different signals within a single cohesive dynamical graph model.