WFTNet: Exploiting Global and Local Periodicity in Long-term Time Series Forecasting

TL;DR

WFTNet is a novel neural network that combines Fourier and wavelet transforms to effectively capture both global and local periodic patterns in long-term time series forecasting, outperforming existing methods.

Contribution

The paper introduces WFTNet, which integrates Fourier and wavelet transforms with a novel PWC mechanism for improved long-term time series forecasting.

Findings

WFTNet outperforms state-of-the-art models on multiple datasets.

The combined use of Fourier and wavelet transforms enhances pattern capturing.

Adaptive weighting improves forecasting accuracy.

Abstract

Recent CNN and Transformer-based models tried to utilize frequency and periodicity information for long-term time series forecasting. However, most existing work is based on Fourier transform, which cannot capture fine-grained and local frequency structure. In this paper, we propose a Wavelet-Fourier Transform Network (WFTNet) for long-term time series forecasting. WFTNet utilizes both Fourier and wavelet transforms to extract comprehensive temporal-frequency information from the signal, where Fourier transform captures the global periodic patterns and wavelet transform captures the local ones. Furthermore, we introduce a Periodicity-Weighted Coefficient (PWC) to adaptively balance the importance of global and local frequency patterns. Extensive experiments on various time series datasets show that WFTNet consistently outperforms other state-of-the-art baseline. Code is available at https://github.com/Hank0626/WFTNet.