Time Adaptive Gaussian Model

TL;DR

The paper introduces the Time Adaptive Gaussian Model (TAGM), a novel approach combining Hidden Markov Models and Gaussian Graphical Models to improve temporal network inference in multivariate time series analysis.

Contribution

It generalizes existing methods by integrating pattern recognition and probabilistic relationship inference, reducing assumptions and enhancing performance.

Findings

Outperforms current methods in temporal network inference.

Effectively clusters data points in time.

Identifies probabilistic and causal relationships.

Abstract

Multivariate time series analysis is becoming an integral part of data analysis pipelines. Understanding the individual time point connections between covariates as well as how these connections change in time is non-trivial. To this aim, we propose a novel method that leverages on Hidden Markov Models and Gaussian Graphical Models -- Time Adaptive Gaussian Model (TAGM). Our model is a generalization of state-of-the-art methods for the inference of temporal graphical models, its formulation leverages on both aspects of these models providing better results than current methods. In particular,it performs pattern recognition by clustering data points in time; and, it finds probabilistic (and possibly causal) relationships among the observed variables. Compared to current methods for temporal network inference, it reduces the basic assumptions while still showing good inference performances.