论文部分内容阅读
本文从自然环境、经济与政策环境、供给、需求等方面设置29个指标构建蔬菜价格预警指标体系,运用均方差分析、相关性分析、主成分分析以及综合选取法来对蔬菜价格预警系统特征指标进行评价,并确定了成本利润率、蔬菜播种面积、城镇居民家庭恩格尔系数、相关替代品价格指数、农村居民对蔬菜需求量、人民币汇率为蔬菜价格预警系统的特征指标,最后建立基于支持向量机SVM(Support Vector Machine)的蔬菜价格预警模型。结果表明,该模型预警准确率高,能够用来进行蔬菜价格预警。
In this paper, we set up 29 indexes to forecast the vegetable price early warning system from the aspects of natural environment, economy and policy environment, supply, demand and so on. By using means of analysis of variance, correlation analysis, principal component analysis and comprehensive selection, And determine the cost margins, sown area of vegetables, Engel coefficient of urban households, the relative price index of alternatives, the demand of rural residents for vegetables, the exchange rate of RMB for the vegetable price warning system characteristics index, and finally establish a support vector machine SVM (Support Vector Machine) vegetable price warning model. The results show that the model has high warning accuracy and can be used for early warning of vegetable prices.