模型变量的选择中国统计教育学会.docVIP

中国粮食供求模型的构建与粮食安全状况的研究 摘要 粮食是人类赖以生存的宝贵资源，是关系国计民生的重要商品。我国是人多地少的发展中国家，要用占世界7％的耕地来养活占世界22％的人口，因此粮食问题，不仅仅是经济问题，更关乎社会的稳定、国家的长治久安。 本模型在总结有关学者对粮食供需问题研究的基础上，结合现阶段我国粮食消费和生产情况，通过搜集数据，分别建立粮食需求和供给模型，并预测未来粮食的供求量，较为系统地分析了粮食供需平衡及粮食安全问题，对我国粮食问题提供借鉴。 首先，本文利用逐步回归方法建立线性回归、残差自回归模型与C-D生产函数、残差自回归模型对可能会影响粮食产量的各个因素进行分析，用R软件对结果进行共线性检验与残差分析，再利用聚类分析方法对粮食产量波动原因做深层次分析，最后得到粮食产量回归方程。通过matlab模拟曲线分别预测四个主要影响因素的值，从来得到未来中国粮食产量。然后，通过整合汇总数据对中国粮食需求总量进行分析，分别建立粮食需求高值模型与低值模型，给出未来几年粮食需求预测值，通过比较未来粮食的产量和需求量的大小，结合实际，研究中国粮食安全状况并给出合理化建议。 关键字: 粮食生产函数 逐步回归 时间序列 残差自回归 多重共线性 聚类分析 粮食需求函数 Abstract Food is a valuable resource for human survival, and peoples livelihood is related to the important commodities. China is a developing countries, using 7% of the worlds arable land to feed 22 percent of the worlds population, so the food problem, is not only an economic issue, but also have something to do with the social and countrys long-term stability. The model summary is based on some scholars’ studies about the food supply and demand, combined with the status of Chinas food consumption and production, through the collection of data by the data from China Statistical Yearbook and China Statistics Bureau Web site , establish the food demand and supply models, and predict future food demand and supply, make more systematic analysis of food supply and demand balance and food security, providing some reference for Chinas food problem. First of all, this paper use the stepwise regression method to establish linear regression, residuals from the regression model, the CD production function, residuals from the regression model may affect the grain yield of the various factors were analyzed with R software testing and the results of linear residual analysis, re-use cluster analysis on the reasons for fluctuations in food production to do in-depth analysis, the final grain yield obtained regression equation. Use Matlab simulation curves to predict four major factors affecting the value, and then calculate Chinas future foo