Nested Spatio-Temporal Time Series Forecasting

Unknown Date

AI Summary

This work addresses the challenge of modeling dynamic spatiotemporal interactions in noisy, non-stationary environments, where existing methods often fail due to their reliance on static spatial priors, leading to systematic biases. To overcome this limitation, we propose a macro-guided nested spatiotemporal forecasting framework. The approach first constructs semantically coherent regional partitions via spectral clustering and then introduces a coarse-to-fine hierarchical predictor that integrates future macroscopic trends with fine-grained historical observations, enabling top-down guidance for high-resolution forecasting. By jointly reasoning across macro and micro scales, our method effectively suppresses noise while preserving long-term dynamics. Extensive experiments on multiple high-dimensional datasets demonstrate substantial improvements over state-of-the-art models, validating the efficacy and robustness of the proposed macro–micro joint inference mechanism.

Abstract

Spatiotemporal forecasting is critical for real-world applications like traffic management, yet capturing reliable interactions remains challenging under noisy and non-stationary conditions. Existing methods primarily rely on historical spatial priors, often failing to account for evolving temporal correlations and suffering from systematic errors. In this work, we propose a nested forecasting framework that couples future macro-level regional trends with micro-level historical observations, enabling top-down guidance from abstract future representations for fine-grained forecasting. Specifically, we employ a spectral clustering-based approach to construct semantically coherent regions, providing both theoretical and empirical evidence that this representation effectively filters systematic noise while preserving essential trends. Building on this, we develop a progressive coarse-to-fine predictor to integrate these representative features into the inference process. This enables the model to leverage trend predictions to anticipate dynamic anomalies, such as periodic offsets, in advance. Furthermore, extensive experiments on multiple high-dimensional datasets demonstrate that our method consistently outperforms state-of-the-art baselines, validating the effectiveness of future macro-guided nested forecasting.

Problem

• spatiotemporal forecasting
• non-stationary conditions
• systematic errors
• temporal correlations
• noisy data

Innovation

• nested forecasting
• spatio-temporal time series
• spectral clustering
• coarse-to-fine prediction
• macro-guided forecasting