Spectral Temporal Graph Neural Network for Multivariate Time-series Forecasting

TL;DR

StemGNN is a novel spectral domain graph neural network that jointly models inter-series and temporal correlations for multivariate time-series forecasting, automatically learning relationships without pre-defined priors.

Contribution

It introduces a spectral domain approach combining GFT and DFT within an end-to-end framework, automatically capturing inter-series correlations and temporal dependencies.

Findings

Outperforms existing methods on ten real-world datasets

Effectively captures complex inter-series and temporal relationships

Automatically learns inter-series correlations from data

Abstract

Multivariate time-series forecasting plays a crucial role in many real-world applications. It is a challenging problem as one needs to consider both intra-series temporal correlations and inter-series correlations simultaneously. Recently, there have been multiple works trying to capture both correlations, but most, if not all of them only capture temporal correlations in the time domain and resort to pre-defined priors as inter-series relationships. In this paper, we propose Spectral Temporal Graph Neural Network (StemGNN) to further improve the accuracy of multivariate time-series forecasting. StemGNN captures inter-series correlations and temporal dependencies \textit{jointly} in the \textit{spectral domain}. It combines Graph Fourier Transform (GFT) which models inter-series correlations and Discrete Fourier Transform (DFT) which models temporal dependencies in an end-to-end framework. After passing through GFT and DFT, the spectral representations hold clear patterns and can be predicted effectively by convolution and sequential learning modules. Moreover, StemGNN learns inter-series correlations automatically from the data without using pre-defined priors. We conduct extensive experiments on ten real-world datasets to demonstrate the effectiveness of StemGNN. Code is available at https://github.com/microsoft/StemGNN/