mbygzh 發(fā)表于 2016-8-12 20:09
謝謝您的解答!但是不知道為什么我用xtivreg y x1 (x2=z),fe 這個命令 結果中沒有z的系數(shù)
建議你先弄明白工具變量法的含義,這個xtivreg默認的是2sls方法,見stata幫助文件xt:
xtivreg offers five different estimators for fitting panel-data models in which some of the right-
hand-side covariates are endogenous. These estimators are two-stage least-squares generalizations of
simple panel-data estimators for exogenous variables. xtivreg with the be option uses the two-
stage least-squares between estimator. xtivreg with the fe option uses the two-stage least-squares
within estimator. xtivreg with the re option uses a two-stage least-squares random-effects estimator.
There are two implementations: G2SLS from Balestra and Varadharajan-Krishnakumar (1987) and
EC2SLS from Baltagi. The Balestra and Varadharajan-Krishnakumar G2SLS is the default because it is
computationally less expensive. Baltagi’s EC2SLS can be obtained by specifying the ec2sls option.
xtivreg with the fd option requests the two-stage least-squares first-differenced estimator.
二階段最小二乘法,第一階段是用內生解釋變量(你這里的X2)對工具變量回歸(z),得到x2的擬合值,然后在第二階段,用這個擬合值作為解釋變量對模型進行回歸的。一般默認的2sls只顯示第二階段的結果,不會顯示z. 你如果想看z的系數(shù),可以在后面加入 r first,即xtivreg y x1 (x2=z), fe r first. 但是其中z的系數(shù)反映的是x2=α+βz+ϵ的回歸結果,這個系數(shù)與y無關