Research on natural gas price forecasting model based on CEEMD-ELM-ARIMA

Author of the article:ZHANG Jinliang, LIU Ziyi, WANG Mingxue

Author's Workplace：School of Economics and Management, North China Electric Power University, Beijing, 102206, China

Key Words：Natural gas price prediction; Extreme Learning Machine; Complementary Ensemble Empirical Mode Decomposition; Hybrid model; Signal processing

Abstract：Natural gas, as a clean energy source, has started to become one of the common energy sources used by many countries in recent years. Natural gas price is an important factor influencing the major business decisions and operational plans made by natural gas companies. The ability to accurately predict the price of natural gas is of great significance to the natural gas industry. Firstly, the natural gas price is broken down using Complementary Ensemble Empirical Mode Decomposition (CEEMD). Secondly, the Extreme Learning Machine (ELM) is used to predict the high-frequency component, and the differential integration of Autoregressive Integrated Moving Average Model (ARIMA) is adopted to predict the low-frequency component. The final prediction result is the sum of the prediction results of each component. The data published by USA Henry Hub Trading Center is used for prediction, and the results of the approach proposed in this paper are compared with other models. The results show that the forecasting model proposed in this paper has higher prediction accuracy.