摘要
基于北京市城镇住户调查数据,采用分位数回归与分解技术,结合核密度估计方法,定量考察了性别、年龄、受教育年限、工作经验等因素对城镇家庭收入差异及其演变的贡献。研究表明:不同收入群体并没有均等地分享经济社会发展的成果;城镇家庭收入差异是"回报效应"和"变量效应"共同作用的结果,不同历史时期两种效应的重要性有所不同;教育机会不均等是导致城镇家庭收入差异的关键因素;就业性质的差异助长了城镇家庭收入差距;不同收入群体间自身素质的差异是影响城镇家庭收入差异的重要因素;工作年限、工作经验、年龄、性别及家庭规模等因素影响力不容忽视。
Based on the data from urban household surveys of Beijing,this paper exploys the quantile regression,decomposition technique and kernel density estimation method to do statistical analysis on the relationships between household income gap and its impact factors.The results show that different income groups do not equally share the fruits of economic and social development;urban household income gap is the result of the "return effect" and "covariate effect",and these two effects' importance vary in different historical periods;unequal educational opportunity is a key factor for urban household income gap;while different kinds of job also contribute to widen it;personal cultivation in different income groups is an important factor,and other factors,such as work experience,age,gender and family size can't be ignored,either.
出处
《统计与信息论坛》
CSSCI
2011年第3期101-107,共7页
Journal of Statistics and Information
基金
国家社会科学基金重点项目<收入分配公平度的统计方法研究--基于收入分配的公平度测算及因素分解>(08ATJ002)
北京市哲学社会科学重点规划项目<北京市中等收入群体动态变迁及提高中等收入者比重路径研究>(10AaJG354)
关键词
收入差异
分位数回归
分位数分解
income gap
quantile regression
quantile decomposition