Comparative Analysis of Global and Local Probabilistic Time Series Forecasting for Contiguous Spatial Demand Regions

2025-09-09

AI Summary

This study systematically compares global versus local probabilistic time-series models for continuous-space traffic demand forecasting, spanning homogeneous synthetic data to highly heterogeneous real-world shared-bike demand scenarios. Method: We propose a global LightGBM framework embedded with station identifiers, integrating temporal features, transportation infrastructure density, and demographic covariates; principal component analysis (PCA) and K-means clustering are employed for dimensionality reduction and structural guidance. Contribution/Results: The method preserves global structural patterns while effectively capturing local heterogeneity, yielding significant improvements in prediction accuracy, prediction interval coverage probability (PICP), and calibration. Experiments demonstrate that our global model consistently outperforms both pure clustering-based and station-level local models across most scenarios—achieving lower mean squared error, narrower prediction intervals, and higher PICP. It establishes a scalable, robust, and interpretable paradigm for probabilistic spatial demand forecasting under high heterogeneity.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Time-Series/Data Streams

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

This study evaluates three probabilistic forecasting strategies using LightGBM: global pooling, cluster-level pooling, and station-level modeling across a range of scenarios, from fully homogeneous simulated data to highly heterogeneous real-world Divvy bike-share demand observed during 2023 to 2024. Clustering was performed using the K-means algorithm applied to principal component analysis transformed covariates, which included time series features, counts of nearby transportation infrastructure, and local demographic characteristics. Forecasting performance was assessed using prediction interval coverage probability (PICP), normalized interval width (PINAW), and the mean squared error (MSE) of the median forecast. The results show that global LightGBM models incorporating station identifiers consistently outperform both cluster-level and station-level models across most scenarios. These global models effectively leverage the full cross-sectional dataset while enabling local adjustments through the station identifier, resulting in superior prediction interval coverage, sharper intervals, and lower forecast errors. In contrast, cluster-based models often suffer from residual within group heterogeneity, leading to degraded accuracy. Station-level models capture fine-grained local dynamics in heterogeneous settings. These findings underscore that global LightGBM models with embedded station identifiers provide a robust, scalable, and computationally efficient framework for transportation demand forecasting. By balancing global structure with local specificity, this approach offers a practical and effective solution for real-world mobility applications.

Problem

• Evaluating probabilistic forecasting strategies for spatial demand regions
• Comparing global, cluster-level, and station-level LightGBM modeling approaches
• Assessing forecasting performance across homogeneous and heterogeneous datasets

Innovation

• Global LightGBM models with station identifiers
• K-means clustering on PCA-transformed spatial covariates
• Probabilistic forecasting using prediction interval metrics