SDGF: Fusing Static and Multi-Scale Dynamic Correlations for Multivariate Time Series Forecasting
This work addresses the problem of accurate forecasting for multivariate time series in domains like finance or weather, but it appears incremental as it builds on existing graph-based methods with multi-scale enhancements.
The paper tackled the challenge of modeling complex multi-scale inter-series correlations in multivariate time series forecasting by proposing the SDGF network, which fuses static and dynamic graphs, and demonstrated its effectiveness through experiments on benchmark datasets.
Inter-series correlations are crucial for accurate multivariate time series forecasting, yet these relationships often exhibit complex dynamics across different temporal scales. Existing methods are limited in modeling these multi-scale dependencies and struggle to capture their intricate and evolving nature. To address this challenge, this paper proposes a novel Static-Dynamic Graph Fusion network (SDGF), whose core lies in capturing multi-scale inter-series correlations through a dual-path graph structure learning approach. Specifically, the model utilizes a static graph based on prior knowledge to anchor long-term, stable dependencies, while concurrently employing Multi-level Wavelet Decomposition to extract multi-scale features for constructing an adaptively learned dynamic graph to capture associations at different scales. We design an attention-gated module to fuse these two complementary sources of information intelligently, and a multi-kernel dilated convolutional network is then used to deepen the understanding of temporal patterns. Comprehensive experiments on multiple widely used real-world benchmark datasets demonstrate the effectiveness of our proposed model.