Saved in:
Bibliographic Details
Main Authors: Mohan, Jeshwanth, Ramsundar, Bharath, Subramanian, Sandya
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.01301
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911296261193728
author Mohan, Jeshwanth
Ramsundar, Bharath
Subramanian, Sandya
author_facet Mohan, Jeshwanth
Ramsundar, Bharath
Subramanian, Sandya
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_01301
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Inferring Dynamic Hidden Graph Structure in Heterogeneous Correlated Time Series
Mohan, Jeshwanth
Ramsundar, Bharath
Subramanian, Sandya
Methodology
Signal Processing
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.
title Inferring Dynamic Hidden Graph Structure in Heterogeneous Correlated Time Series
topic Methodology
Signal Processing
url https://arxiv.org/abs/2512.01301