动态面板模型中个体效应设定方法与检验  

Individual Effect Method Setting and Testing in Dynamic Panel Model

作　　者：尹康 洪丽[2,3] Yin Kang;Hong Li(School of Economics and Trade,Hubei University of Economics,Wuhan 430205,China;a.School of Political Science and Public Administration,Wuhan 430072,China;Center for Social Security Study,Wuhan University,Wuhan 430072,China)

机构地区：[1]湖北经济学院经济与贸易学院,武汉430205 [2]武汉大学政治与公共管理学院,武汉430072 [3]武汉大学社会保障研究中心,武汉430072

出　　处：《统计与决策》2023年第8期24-29,共6页Statistics & Decision

基　　金：教育部人文社会科学研究青年基金项目(16YJC790027);“中央高校基本科研业务费专项资金”武汉大学自主科研项目(人文社会科学)(413000037)。

摘　　要：文章对比分析了动态面板模型中个体效应被设定为固定效应和随机效应时,差分GMM估计量和系统GMM估计量所具有的不同统计特征,在此基础上构造了基于差分GMM估计和系统GMM估计的Hausman检验来识别模型中的个体效应,并通过Bootstrap抽样模拟计算了Hausman检验的水平和功效函数。模拟结果表明,Hausman检验比传统J检验(Sargan-Hansen检验)有更小的水平扭曲和更高的检验功效。最后将Hausman检验方法用于跨国别的收入水平与民主化关系研究,检验结果表明,差分GMM更适用于跨国别的动态面板模型估计。This paper compares and analyzes the different statistical characteristics of the difference GMM estimator and the system GMM estimator when the individual effect is set as fixed effect and random effect in the dynamic panel model.On this basis,the paper constructs a Hausman test based on difference GMM estimation and system GMM estimation to identify individual effects in the model,and calculates the size and power function of Hausman test by Bootstrap sampling simulation.The simulation results show that the Hausman test has less size distortion and higher test power than traditional J test(Sargan-Hansen test).Finally,the paper applies Hausman test to the study on the relationship between income and democratization in different countries.The test results show that difference GMM is more suitable for the transnational dynamic panel model estimation.

关 键 词：系统GMM 差分GMM HAUSMAN检验 自助法 

分 类 号：O212.1[理学—概率论与数理统计]