Long-term Time Series Forecasting based on Decomposition and Neural Ordinary Differential Equations
This work addresses long-term forecasting problems in domains like finance and healthcare, but it is incremental as it builds on existing neural ODE and decomposition methods.
The paper tackles the challenge of long-term time series forecasting by proposing LTSF-DNODE, which combines linear ordinary differential equations with time series decomposition to address limitations in existing Transformer-based and Linear-based models, and shows it outperforms baselines on various real-world datasets.
Long-term time series forecasting (LTSF) is a challenging task that has been investigated in various domains such as finance investment, health care, traffic, and weather forecasting. In recent years, Linear-based LTSF models showed better performance, pointing out the problem of Transformer-based approaches causing temporal information loss. However, Linear-based approach has also limitations that the model is too simple to comprehensively exploit the characteristics of the dataset. To solve these limitations, we propose LTSF-DNODE, which applies a model based on linear ordinary differential equations (ODEs) and a time series decomposition method according to data statistical characteristics. We show that LTSF-DNODE outperforms the baselines on various real-world datasets. In addition, for each dataset, we explore the impacts of regularization in the neural ordinary differential equation (NODE) framework.