Forecasting Multivariate Urban Data via Decomposition and Spatio-Temporal Graph Analysis
It addresses forecasting challenges for urban data applications, such as electricity demand and air pollution, with incremental improvements over existing methods.
This paper tackles the problem of long-term forecasting of multivariate urban data by proposing DST, a model that integrates graph attention and temporal convolution with decomposition-based preprocessing, achieving an average improvement of 2.89% to 9.10% in accuracy over state-of-the-art models.
Long-term forecasting of multivariate urban data poses a significant challenge due to the complex spatiotemporal dependencies inherent in such datasets. This paper presents DST, a novel multivariate time-series forecasting model that integrates graph attention and temporal convolution within a Graph Neural Network (GNN) to effectively capture spatial and temporal dependencies, respectively. To enhance model performance, we apply a decomposition-based preprocessing step that isolates trend, seasonal, and residual components of the time series, enabling the learning of distinct graph structures for different time-series components. Extensive experiments on real-world urban datasets, including electricity demand, weather metrics, carbon intensity, and air pollution, demonstrate the effectiveness of DST across a range of forecast horizons, from several days to one month. Specifically, our approach achieves an average improvement of 2.89% to 9.10% in long-term forecasting accuracy over state-of-the-art time-series forecasting models.