Learning Leading Indicators for Time Series Predictions

TL;DR

This paper introduces new methods for forecasting multiple time series by identifying key leading indicators and uncovering system structures, improving prediction accuracy and interpretability.

Contribution

It proposes two novel approaches for learning sparse VAR models with leading indicators, including clustering based on Granger-causality, advancing time series forecasting techniques.

Findings

Methods outperform state-of-the-art sparse VAR models

Effective in discovering meaningful Granger-causality structures

Improves forecasting accuracy across diverse datasets

Abstract

We consider the problem of learning models for forecasting multiple time-series systems together with discovering the leading indicators that serve as good predictors for the system. We model the systems by linear vector autoregressive models (VAR) and link the discovery of leading indicators to inferring sparse graphs of Granger-causality. We propose new problem formulations and develop two new methods to learn such models, gradually increasing the complexity of assumptions and approaches. While the first method assumes common structures across the whole system, our second method uncovers model clusters based on the Granger-causality and leading indicators together with learning the model parameters. We study the performance of our methods on a comprehensive set of experiments and confirm their efficacy and their advantages over state-of-the-art sparse VAR and graphical Granger learning methods.