【ESG】初步回归结果（Hausman检验+固定效应）

最新推荐文章于 2023-02-25 10:39:41 发布

Huckin119

最新推荐文章于 2023-02-25 10:39:41 发布

阅读量2.3k

点赞数

文章标签：开发语言 Powered by 金山文档

本文链接：https://blog.csdn.net/weixin_70420642/article/details/129107798

版权

这篇博客探讨了在环境、社会和治理（ESG）数据分析中，如何运用Hausman检验来确定模型的适用性，并详细阐述了固定效应模型在ESG研究中的应用，旨在提供一种更精确的数据解释方法。

摘要由CSDN通过智能技术生成

. hausman FE RE, constant sigmamore

                 ---- Coefficients ----
             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))
             |       FE           RE         Difference          S.E.
-------------+----------------------------------------------------------------
      Escore |    -.035016    -.1463142        .1112983        .0523231
      Sscore |   -.0818201    -.1024889        .0206688        .0304015
      Gscore |     .132224     .0253786        .1068454        .0242963
       _cons |    1.864511     2.748634       -.8841227        .1916242
------------------------------------------------------------------------------
                           b = consistent under Ho and Ha; obtained from xtreg
            B = inconsistent under Ha, efficient under Ho; obtained from xtreg

    Test:  Ho:  difference in coefficients not systematic

                  chi2(4) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                          =       24.21
                Prob>chi2 =      0.0001  //选用固定效应模型
                (V_b-V_B is not positive definite)

*进行固定效应回归，控制到个体层面
reghdfe TobinQA Escore Sscore Gscore , absorb(year stkcd) vce(cluster stkcd)
(dropped 652 singleton observations)
(MWFE estimator converged in 8 iterations)

HDFE Linear regression                            Number of obs   =     18,635
Absorbing 2 HDFE groups                           F(   3,   3685) =       3.54
Statistics robust to heteroskedasticity           Prob > F        =     0.0141
                                                  R-squared       =     0.3082