陕西省畜牧业总产值组合预测模型研究基于Shapley值  

Prediction of the Gross Output Value of Animal Husbandry in Shanxi Province Based on Shapley Combination Forecasting Model

作　　者：邢书宝[1] 延秀娟[1] XING Shu-bao;YAN Xiu-juan(School of Management,Xi'an University of Science and Technology,Xi'an 710054,China)

机构地区：[1]西安科技大学管理学院,陕西西安710054

出　　处：《数学的实践与认识》2020年第16期34-42,共9页Mathematics in Practice and Theory

基　　金：国家社科基金项目(13BJY098)农村剩余劳动力动态估算方法研究及应用;陕西省社会科学基金项目(2014D38)陕西省农村剩余劳动力估算及预测研究;陕西省教育厅项目(12JK0068)。

摘　　要：依据组合预测模型理论,利用2006-2016年陕西省畜牧业总产值的数据,分别建立了畜牧业产值的一元回归模型、BP神经网络模型、三次指数平滑模型和基于Shapley值的组合预测模型,得到各模型的预测精度分别为7.38%、4.46%、3.50%和2.90%,可见组合模型的拟合精度低于各单一模型,拟合效果好.为此,利用所得模型对2017-2025年陕西省畜牧业总产值进行了预测.Based on the theory of combined forecasting model,this paper builds the single regression model,BP neural network model,cubic exponential smoothing model and the combined forecasting mode by using the 2006-2016 data of the gross output value of animal husbandry output value in Shanxi Province.The forecasting accuracy of the combined forecasting model was 2.90%,which is lower than 7.38%,4.46% and 3.50% of the single prediction models such as single regression model,BP neural network model and cubic exponential smoothing model.It is suggested that the combination fitting effect is good.Therefore,Therefore,the combined forecasting model is used to predict the output value of animal husbandry in Shanxi province from 2017 to 2025.

关 键 词：畜牧业 总产值 BP神经网络 组合预测 指数平滑 

分 类 号：F326.3[经济管理—产业经济] F224