灰色关联分位评价方法与多维相对贫困测度  

Grey Incidence Quantile Evaluation Method and Multidimensional Relative Poverty Measurement

作　　者：刘震 谢玉梅 丁松 Liu Zhen;Xie Yumei;Ding Song(School of Digital Economy and Management,Wuxi University,Wuxi Jiangsu 214105,China;School of Business,Jiangnan University,Wuxi Jiangsu 214122,China;School of Economics,Zhejiang University of Finance and Economics,Hangzhou 310018,China)

机构地区：[1]无锡学院数字经济与管理学院,江苏无锡214105 [2]江南大学商学院,江苏无锡214122 [3]浙江财经大学经济学院,杭州310018

出　　处：《统计与决策》2023年第21期40-45,共6页Statistics & Decision

基　　金：国家自然科学基金资助项目(72001093,71901191);国家社会科学基金重大专项(18VSJ098);教育部人文社会科学研究青年基金项目(17YJC790101)。

摘　　要：为了提升乡村振兴阶段相对贫困人口的识别准确性,文章聚焦贫困的多维属性,利用灰色关联理论构建了一类新的具有动态特征的多维贫困识别方法。该方法将被测主体视为空间中的几何形状,利用几何特征对比替代传统方法中的双重临界值,降低了临界值选取对识别结果的主观影响,以分位数指标为基础实现了对相对贫困的动态识别。同时拓展了灰色关联评价模型,讨论了识别结果的加总与分解,建立了多维贫困发生率、多维贫困分布指数与多维贫困深度指数。最后利用中国家庭追踪调查(CFPS)2018年的数据对中国多维相对贫困情况进行分析,结果表明,我国农村的多维贫困发生率远高于城市,但是贫困分布和贫困深度差距不大,各贫困维度的贡献率在不同区域和城乡范围内表现出不同的特征。In order to improve the identification accuracy of the relative poverty population in the rural revitalization stage,this paper focuses on the multidimensional attributes of poverty and constructs a new multidimensional poverty identification method with dynamic characteristics by using grey incidence theory.In this method,the measured subject is regarded as the geometry in space,and the geometric feature comparison is used to replace the double critical value in the traditional method,so that the subjective impact between the selection of the critical value and the recognition results is reduced,and the dynamic recognition of relative poverty based on the quantile index is realized.What’s more,the grey incidence evaluation system is improved,which discusses the aggregation and decomposition of the identification results,including the multidimensional poverty index,multidimensional poverty distribution index,and multidimensional poverty depth index.Finally,an experimental study on measuring and analyzing China’s multidimensional poverty level is implemented based on the Chinese Family Panel Studies(CFPS)data in 2018.The results demonstrate that the multidimensional poverty incidence in rural area is much higher than that in city,but that the poverty distribution and poverty depth between rural area and cities are small,and that the contribution rate of each poverty dimension shows different characteristics in different regions.

关 键 词：灰色系统 灰色关联 多维贫困 相对贫困 

分 类 号：F224.9[经济管理—国民经济]