组合分布在我国居民收入分布拟合中的应用研究

发布时间：2018-03-23 00:34

本文选题：CHNS　切入点：居民收入　出处：《统计研究》2017年06期 　论文类型：期刊论文

【摘要】：本文使用CHNS家庭微观调查数据研究发现,我国城乡居民家庭人均收入对数呈现明显左偏特征。为了适应这种需要,根据组合分布函数的构造思想,首次提出使用反向帕累托-对数正态组合分布对居民收入数据进行建模,并使用极大似然估计方法对参数进行估计。然后,使用变量替换方法导出我国城乡居民收入原始数据的确切分布,并在此基础上,测算出我国城乡居民收入的基尼系数。实证研究发现,农村居民家庭人均收入基尼系数高于城镇居民,就基尼系数发展态势而言,我国城乡居民家庭收入基尼系数在2007—2009年间发生了转变,由之前的持续增加转为下降。
[Abstract]:In this paper, we find that the logarithm of per capita income of urban and rural households in China is obviously left-biased by using CHNS family microscopic survey data. In order to meet this need, according to the construction idea of combinatorial distribution function, It is the first time to use reverse Pareto-logarithmic normal combination distribution to model the resident income data, and to estimate the parameters by using the maximum likelihood estimation method. The exact distribution of the original income data of urban and rural residents in China is derived by using variable substitution method, and on this basis, the Gini coefficient of income of urban and rural residents in China is calculated. The Gini coefficient of per capita income of rural households is higher than that of urban residents. In terms of the development trend of Gini coefficient, the Gini coefficient of household income of urban and rural residents in China has changed from the previous continuous increase to decline from 2007 to 2009.
【作者单位】：中南财经政法大学统计与数学学院;
【基金】：国家社会科学基金项目“中国农村贫困测度与识别研究”(16BTJ005) 中南财经政法大学博士研究生重点科研创新项目“基于半逻辑分布的可靠性统计推断与应用研究”(2015BZ1302)资助
【分类号】：F124.7;F224

【相似文献】