Non-Stationary Time Series Forecasting Based on Fourier Analysis and Cross Attention Mechanism

TL;DR

This paper introduces AEFIN, a novel deep learning framework that combines Fourier analysis and cross-attention to improve forecasting of non-stationary time series data, outperforming existing models in accuracy and robustness.

Contribution

The paper proposes a new framework, AEFIN, integrating Fourier analysis, cross-attention, and a specialized loss function to better model non-stationary time series.

Findings

AEFIN outperforms common models in mean square and absolute error.

The framework effectively captures seasonal and trend patterns in unstable components.

AEFIN demonstrates strong forecasting capabilities under non-stationary conditions.

Abstract

Time series forecasting has important applications in financial analysis, weather forecasting, and traffic management. However, existing deep learning models are limited in processing non-stationary time series data because they cannot effectively capture the statistical characteristics that change over time. To address this problem, this paper proposes a new framework, AEFIN, which enhances the information sharing ability between stable and unstable components by introducing a cross-attention mechanism, and combines Fourier analysis networks with MLP to deeply explore the seasonal patterns and trend characteristics in unstable components. In addition, we design a new loss function that combines time-domain stability constraints, time-domain instability constraints, and frequency-domain stability constraints to improve the accuracy and robustness of forecasting. Experimental results show that AEFIN outperforms the most common models in terms of mean square error and mean absolute error, especially under non-stationary data conditions, and shows excellent forecasting capabilities. This paper provides an innovative solution for the modeling and forecasting of non-stationary time series data, and contributes to the research of deep learning for complex time series.