论文部分内容阅读
利用CFPS 2010年数据,着重检验了父母职业地位对子女职业的直接影响。在数据处理方面,为了得到系数及标准误的无偏估计,对于抽样数据的缺失值,采用了贝叶斯多重插值法对因变量与自变量进行填补。在统计模型选取方面,利用OLS、工具变量法及两阶段最小二乘法(2SLS)对估计偏倚进行校正。为了选择最优权重矩阵,采用广义矩估计(GMM)及迭代矩估计(IGMM)来检验各因素特别是父亲职业对子女职业效应的直接影响及地区差异。为了避免样本选择带来的估计偏倚,引入Heckman选择模型对工具变量法及广义矩估计等模型的估计结果进行稳健性检验。研究发现父辈职业地位高的,其子女的职业地位也相应高,这从程序上说明我国阶层固化作用的存在。通过Heckman选择模型,研究还新发现父亲职业地位对子女选择某职业与否以及对子女现有职业的发展都有显著影响。此外,这种影响也表现为地域差异,经济水平越发达或文化越开放,父亲职业的影响力就越强;但个体受教育程度是影响其职业地位的主导因素。
Using the CFPS 2010 data, we examined the direct impact of parental occupational status on children’s careers. In data processing, in order to get unbiased estimation of coefficients and standard errors, Bayesian multiple interpolation method is used to fill dependent variables and independent variables for the missing values of sampling data. In the aspect of statistical model selection, OLS, instrumental variable method and two-stage least squares (2SLS) are used to correct the estimation bias. In order to choose the optimal weight matrix, GMM and IGMM were used to test the direct effects of various factors, especially the fathers’ occupational effects on children’s occupational effects and regional differences. In order to avoid the estimation bias caused by the sample selection, Heckman selection model is introduced to test the robustness of the estimation results of the model such as the tool variable method and the generalized moment estimation. Study found that his father’s high occupation status, their children’s occupational status is also high, which shows that the process of curing the role of our class exist. Through Heckman’s selection model, the study also found that father’s professional status has a significant impact on children’s choice of occupation or not and on the development of their children’s existing occupations. In addition, this kind of influence is also manifested as geographical differences. The more developed the economy or the more open the culture, the stronger the influence of the father’s profession. However, the education level of the individual is the dominant factor affecting his occupational status.