Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • #31
    Hi everyone,
    I do not see any endogeneity test statistic being reported for xtivreg, which is available in xtivreg2 (I guess). I am not sure if the statistics we get for xtivreg FE model with the first stage results being specified reports endogeneity stats. Kindly clarify on this.
    Regards,

    Comment


    • #32
      Originally posted by Alex Mai View Post

      Dear Jeff,

      I tried to test the endogeneity of a variable under random-effects panel regression, using the method you mentioned. However, I do not understand why the number of exogenous variables in -xtreg y1 y2 v2hat z1 z2 ... zM, re vce(cluster id)- is smaller than that in -reg y2 z1 z2 ... zL-, (M<L).
      Does it mean that I should randomly delete several exogenous variables in -xtreg-?

      Thank you very much!
      Hello all,
      I would like to know whether we need to exclude some exogenous variables in the final outcome equation (in other words include more controls in the first stage regression where we calculate the residuals)? If that is the case, how do we make the choice of the exogenous control variables?

      Comment

      Working...
      X