摘要
本文利用"中国营养和健康调查"家庭平均收入的模块数据,采用核密度估计、累积分布函数、洛伦茨曲线等方法刻画我国城乡居民收入分布的动态演进,结果表明,大多数家庭都分享了我国经济快速增长带来的成果,其中高收入阶层比低收入阶层更多地分享了经济成果,城市居民比农村居民更多地分享了经济成果。进一步地,非参数假设检验和相对偏差估计结果表明,N.C.Kakwani的三参数函数和简易公式的基尼系数估计精度较高,广义Logistic分布函数是估计城乡居民收入分布的可行选择。
Utilizing panel data of household incomes extracted from the China Nutrition and Health Survey, this paper depicts the dynamic of the urban-rural income distribution by applying kernel density estimation, cumulative distribution function as well as Lorenz curve. As the empirical result shows, most households had shared the fruits of rapid economic growth, particularly, the higher-income family shared more than lower-income family, and urban family shared more than rural family. Furthermore, the results of nonparametric test and relative bias indicate that, the N. C. Kakwani's three-parameter model and the concise formula are both ideal choices for the calculation of Gini coefficients, and the generalized logistic distribution is an alternative choice for estimating the income distribution.
出处
《统计研究》
CSSCI
北大核心
2009年第12期32-40,共9页
Statistical Research
关键词
分布动态
非参数检验
基尼系数
Dynamics of Income Distribution
Nonparametric Test
Gini Coefficients