Time Series Extrinsic Regression

TL;DR

This paper introduces Time Series Extrinsic Regression (TSER), a new task that predicts continuous variables from time series, benchmarks existing algorithms on a new dataset collection, and finds that adapted TSC algorithms like Rocket perform best.

Contribution

It defines TSER as a novel regression task related to TSC and TSF, and provides a comprehensive benchmark of algorithms on a new dataset archive.

Findings

Rocket adapted for regression outperforms other algorithms

Current models need significant improvement in accuracy

Research prospects are promising for advancing TSER methods

Abstract

This paper studies Time Series Extrinsic Regression (TSER): a regression task of which the aim is to learn the relationship between a time series and a continuous scalar variable; a task closely related to time series classification (TSC), which aims to learn the relationship between a time series and a categorical class label. This task generalizes time series forecasting (TSF), relaxing the requirement that the value predicted be a future value of the input series or primarily depend on more recent values. In this paper, we motivate and study this task, and benchmark existing solutions and adaptations of TSC algorithms on a novel archive of 19 TSER datasets which we have assembled. Our results show that the state-of-the-art TSC algorithm Rocket, when adapted for regression, achieves the highest overall accuracy compared to adaptations of other TSC algorithms and state-of-the-art machine learning (ML) algorithms such as XGBoost, Random Forest and Support Vector Regression. More importantly, we show that much research is needed in this field to improve the accuracy of ML models. We also find evidence that further research has excellent prospects of improving upon these straightforward baselines.