Energy Commodity Price Volatility and Forecasting in China: Evidence from the Pre- and Post-COVID-19 Periods

Abstract

This study examines changes in energy commodity price behaviour and forecasting performance before and after the COVID-19 pandemic, focusing on coal prices (CP), uranium prices (UP), and crude oil price volatility (COPVOL). Using data from January 2018 to June 2022, the study applies Autoregressive Integrated Moving Average (ARIMA), Single Exponential Smoothing (SES), and K-Nearest Neighbors (K-NN) models. Forecasting performance is evaluated using MAE, MAPE, and RMSE. The findings indicate changes in the level and variability of the examined series following COVID-19. Based on the reported forecasting errors, ARIMA generally outperforms SES and K-NN across the selected energy series. The findings highlight the importance of reliable forecasting for energy-market risk assessment, investment planning, and policy decision-making under changing market conditions.