FEDEL: Frequency Enhanced Decomposition and Expansion Learning for Long-Term Time Series Forecasting

Long-term Time Series Forecasting (LTSF) is widely used in various fields, for example, power planning. LTSF requires models to capture subtle long-term dependencies in time series effectively. However, several challenges hinder the predictive performance of existing models, including the inability to exploit the correlation dependencies in time series fully, the difficulty in decoupling the complex cycles of time series in the time domain, and the error accumulation of iterative multi-step prediction. To address these issues, we design a Frequency Enhanced Decomposition and Expansion Learning (FEDEL) model for LTSF. The model has a linear complexity with three distinguishing features: (i) an extensive capacity depth regime that can effectively capture complex dependencies in long-term time series, (ii) decoupling of complex cycles using sparse representations of time series in the frequency domain, (iii) a direct multi-step prediction strategy to generate the prediction series, which can improve the prediction speed and avoid error accumulation. We have conducted extensive experiments on eight real-world large-scale datasets. The experimental results demonstrate that the FEDEL model performs significantly better than traditional methods and outperforms the current SOTA model in the field of LTSF in most cases.