基于时间序列模型的短期中国能源消费量预测研究

Short-Term Energy Consumption Forecasting in China Based on Time Series Model

摘要: 能源的使用与社会生产生活有着密切的联系，对此能源消费量的准确预测对于制定能源生产和消费规划以及保持中国经济健康、持续发展具有极其重要的理论与现实意义。本文采用ARIMA模型，利用中国1990~2019年的能源消费量进行时间序列建模，2020~2021年能源消费量数据进行测试。通过相关的检验和分析，疏系数ARIMA((5),2,1)模型的拟合效果较好，对比2020和2021年真实数据发现预测平均误差仅为0.835%，因此利用该模型对中国未来三年的能源消费量进行预测。预测结果表明，未来三年中国能源消费量仍然增加，但是增长速度放缓，符合中国当前低碳经济发展背景。

Abstract: The use of energy is closely related to social production and social life, the accurate prediction of energy consumption for the development of energy production and consumption planning, keeping China’s economy healthy and sustainable development has the extremely important theoretical and practical significance. In this paper, ARIMA model is adopted to conduct time series modeling with China’s energy consumption from 1990 to 2019, and the data of energy consumption from 2020 to 2021 are used for testing. Through relevant testing and analysis, ARIMA((5),2,1) model has a good fitting effect. Compared with the real data of 2020 and 2021, it is found that the average error of prediction is only 0.835%. Therefore, this model is used to forecast the energy consumption of China in the next three years. The prediction results show that in the next three years, China’s energy consumption will still increase, but the growth rate will slow down, which is in line with China’s current low-carbon economic development background.