Forecasting Coal Freight Rates: An Empirical Comparison of sARIMA and Holt-Winters Approaches

Coal remains one of China's primary energy sources. Road freight prices for coal serve as a "barometer" of the coal market, playing a vital role in cost control for transport enterprises and in broader market regulation. Using historical data from the National Road Freight Market Monitoring System, this study focuses on daily coal road freight rates from 2021 to 2024 in four major coal-producing provinces: Shanxi, Shaanxi, Inner Mongolia, and Xinjiang. Monthly freight rates were derived from daily data, with the 2021–2023 data used as the training set and the 2024 data as the test set for model validation. After a comparative evaluation of forecasting methods, the Seasonal ARIMA (sARIMA) and Holt-Winters exponential smoothing models were selected for their simplicity, accuracy, and applicability to freight pricing scenarios. Forecasting was conducted separately for intra-provincial and inter-provincial transport and across different transport distance categories. Model performance was assessed using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE). The results indicate that the Holt-Winters model is generally preferred, with the sARIMA model serving as a supplementary benchmark when necessary.