基于改进AF法的中国多维相对贫困测度研究  

A Study on Multidimensional Relative Poverty Measurement in China Based on Modified AF Method

作　　者：张英杰 苏宇楠[1] 蔺子航 ZHANG Yingjie;SU Yunan;LIN Zihang(School of science,Minzu University of China,Beijing 100081,China)

机构地区：[1]中央民族大学理学院,北京100081

出　　处：《中央民族大学学报（自然科学版）》2024年第3期73-80,共8页Journal of Minzu University of China(Natural Sciences Edition)

摘　　要：本文采用变异系数修正AHP的改进AF法,基于CFPS数据库运用“双阈值法”,对我国相对贫困问题进行识别和分解,并利用分位数回归讨论高低分位水平下各指标对相对贫困的影响程度。研究表明:第一,我国多维相对贫困的发生率整体呈下降趋势,多维相对贫困主要集中在农村地区;第二,多维相对贫困群体在生活水平、人均家庭收入和儿童失学率遭受的相对剥夺最突出,前两者对相对贫困呈负向影响,儿童失学率对相对贫困呈正向影响;第三,各指标对不同相对贫困水平下的人群影响是不同的。债务对于贫困指标数最多的群体影响最大,文娱支出更能反映小康小富群体的相对贫困情况,人均纯收入、人均住房价值和金融产品更能体现生活条件较好家庭的相对贫困状态。In this paper,the coefficient of variation modified AHP index weighting method was used to improve AF method,and the"double threshold method"was used based on CFPS database to i⁃dentify and decompose China's relative poverty,and quantile regression was used to discuss the in⁃fluence degree of each index on relative poverty at high and low quantile levels.The results show that:First,the incidence of multi⁃dimensional relative poverty in our country shows a downward trend,and the multi⁃dimensional relative poverty is mainly concentrated in rural areas;Secondly,multidimensional relative poor groups suffer the most significant relative deprivation in terms of living standards,per capita family income and child dropout rate.The former two have a negative impact on relative poverty,while the child dropout rate has a positive impact on relative poverty.Third,in⁃dicators have different impacts on people at different levels of relative poverty.Debt has the greatest impact on the groups with the largest number of poverty indicators.Entertainment expenditure can better reflect the relative poverty of well⁃off small and rich groups,and per capita net income,per capita housing value and financial products can better reflect the relative poverty of families with bet⁃ter living conditions.

关 键 词：多维相对贫困指数 修正AHP 改进AF法 

分 类 号：O212[理学—概率论与数理统计]