Interpretable PM2.5 Forecasting for Urban Air Quality: A Comparative Study of Operational Time-Series Models

TL;DR

This study compares lightweight, interpretable models like SARIMAX, Facebook Prophet, and NeuralProphet for hourly PM2.5 forecasting in Beijing, demonstrating that simpler models can achieve competitive accuracy with less computational demand.

Contribution

It introduces a leakage-aware forecasting workflow and evaluates the practical deployment of three time-series models under adaptive regimes for urban air quality prediction.

Findings

Facebook Prophet achieved the best accuracy under walk-forward refitting.

Residual correction improved SARIMAX performance in frozen-model regime.

NeuralProphet was less accurate and stable compared to other models.

Abstract

Accurate short-term air-quality forecasting is essential for public health protection and urban management, yet many recent forecasting frameworks rely on complex, data-intensive, and computationally demanding models. This study investigates whether lightweight and interpretable forecasting approaches can provide competitive performance for hourly PM2.5 prediction in Beijing, China. Using multi-year pollutant and meteorological time-series data, we developed a leakage-aware forecasting workflow that combined chronological data partitioning, preprocessing, feature selection, and exogenous-driver modeling under the Perfect Prognosis setting. Three forecasting families were evaluated: SARIMAX, Facebook Prophet, and NeuralProphet. To assess practical deployment behavior, the models were tested under two adaptive regimes: weekly walk-forward refitting and frozen forecasting with online residual correction. Results showed clear differences in both predictive accuracy and computational efficiency. Under walk-forward refitting, Facebook Prophet achieved the strongest completed performance, with an MAE of $37.61$ and an RMSE of $50.10$, while also requiring substantially less execution time than NeuralProphet. In the frozen-model regime, online residual correction improved Facebook Prophet and SARIMAX, with corrected SARIMAX yielding the lowest overall error (MAE $32.50$; RMSE $46.85$). NeuralProphet remained less accurate and less stable across both regimes, and residual correction did not improve its forecasts. Notably, corrected Facebook Prophet reached nearly the same error as its walk-forward counterpart while reducing runtime from $15$ min $21.91$ sec to $46.60$ sec. These findings show that lightweight additive forecasting strategies can remain highly competitive for urban air-quality prediction, offering a practical balance between accuracy, interpretability, ...