代际研究中终生收入估算偏差问题探讨  

Discussion on Estimation Bias of Lifetime Income in Intergenerational Research

作　　者：苏宇楠[1] 马敬 Su Yunan;Ma Jing(College of Science,Minzu University of China,2.School of Mathematics and Statistics,Beijing Institute of Technology,Beijing 100081,China)

机构地区：[1]中央民族大学理学院 [2]北京理工大学数学与统计学院,北京100081

出　　处：《统计与决策》2020年第8期10-15,共6页Statistics & Decision

基　　金：国家社会科学基金青年项目(16CTJ011)。

摘　　要：在我国代际收入流动性问题研究中,微观数据分析常用父代40岁左右的单年收入替代其终生收入,由此导致代际收入弹性估计偏大问题。文章假定微观个体观测值服从对数正态分布并进行建模,对个体终生收入做出统计预测。模拟得到学历为硕士人群的工资年增长率是本科人群的1.06倍,收入达到峰值的时间则晚了约8年,对全部人群父代的终生收入用其40岁左右收入逼近的思路可依据父代教育水平进行再次细分。In the research of intergenerational income mobility in China,the single yearly income of the father aged around40 is often used to replace his lifetime income in micro data analysis,which leads to the problem of a little too large elasticity estimation of intergenerational income.This paper assumes that the micro individuals observation values follow a logarithmic normal distribution and are modeled to make a statistical prediction of individuals’lifetime income.The simulation shows that the annual wage growth rate of the people with a master’s degree is 1.06 times that of the people with a bachelor’s degree,and the time when the income reaches its peak is about 8 years later;the lifetime income of the whole population of the father generation,if approximated by their income at the age of around 40,can be subdivided again according to the education level of the father generation.

关 键 词：终生收入 估计偏误 对数正态分布 蒙特卡洛模拟 

分 类 号：O202[理学—数学]