摘要
煤炭需求是一个复杂的非线形系统,其发展变化具有增长性和波动性,科学地预测煤炭需求量对于正确制定煤炭工业发展规划和优化配置煤炭资源具有重要意义。本文在对我国煤炭消费历史数据分析的基础上,构建了一元回归模型、灰色预测的GM(1,1)模型和三次指数平滑模型3个单一预测模型;据此,基于二次规划法和Shapley值权重分配法分别构建了组合预测模型。实证研究结果表明,组合模型预测效果均优于单一模型,可以作为煤炭需求预测的有效工具;基于这一组合模型对我国2013—2020年煤炭需求量进行了预测,为制定未来煤炭需求发展规划提供了依据。
Coal demand is a complex non-linear system with growth and volatility. To forecast scientifically the coal demand is very important for formulating coal industry development plan and allocating optimally coal resources. In the paper,three single coal demand forecasting models such as single variable linear regression model GM（ 1,1） model,l and cubic exponential smoothing method are set up based on on historical data of Chinese coal consumption. According to the single models,combination forecasting models for coal demand are respectively constructed using v the quadratic programming and Shapley value method. The results show that prediction accuracy of the combination forecasting model is better than that of three single models. Finally,Using the combination forecasting model,we predict China＇s coal demand from 2013 to 2020,the predict result could provide the decision basis for the future development planning of Chin＇s coal demand.
出处
《西安科技大学学报》
CAS
2014年第5期553-557,共5页
Journal of Xi’an University of Science and Technology
基金
陕西省社科基金项目(2014D38)
陕西省教育厅科学研究计划(12JK0068)
