农产品质量安全风险评估与预警模型:以山东省蔬菜出口示范基地为例

Agricultural food quality safety evaluation and risk early-warning model:Illustrated by an example of export vegetable sites in Shandong province

针对农产品安全风险评估和预警中涉及的语言信息和面板数据,提出了一种基于多分类离散选择模型的评估分析方法.在建立农产品安全风险评估指标体系的基础上,采用语言信息处理方法,将呈现出面板数据特征的评价信息进行有效集结.通过因子分析法提取农产品安全风险评估的潜在变量,建立有序多分类离散选择模型,给出了风险判别分析的计算步骤.结合山东省出口蔬菜备案基地评价数据进行了实证研究.研究表明以上方法可以针对农产品质量安全风险评估的特点,有效地将语言评价信息进行集结和利用,解决了农产品安全风险管理中风险级别有"序"的问题,该方法将促进良好农业操作规范的推广和应用,实现农产品安全风险预警管理的目标.

Based on the linguistic information and the panel data in agricultural food quality risk evaluation and early-warning management,this paper constructed an analysis approach based on multi-classification discrete choice model.On the basis of building an agricultural food safety risk evaluation index system,applying the linguistic information processing method,the panel valuation data aggregated to obtain the integrated evaluation values.The potential variables are extracted by factor analysis method, an ordered logistic model built,and the calculation analysis steps are given.This research is illustrated by an example of export vegetable sites which are put on records in Shandong province.The research indicates that the linguistic information can be effectively aggregated with less lost by this method,the issues that the classes of agricultural food quality risk are ordinal,and approach will promote the extension and application in GAP,and realize the target of agricultural product safety.