Dear Statalist,
I would like to estimate the following equation:
I am concerned about two endogeneity problems:
Put differently, I have two outcomes Y1 and Y2, where Y1 may amongst others depend on Y2, so an alternative to the above IV could be a Simultaneous Equations Model of the form:
Equation (2) is basically the First Stage and Equation (1) the Second Stage of the IV estimation suggested above. However, the problem is: Whether Y1 and Y2 are observed, does also depend on many of the same regressors Xb, Xc, Z2, which I think would call for a Heckman selection model. But how do I combine SEM and Heckman model? If I simply add the selection equation as 3rd equation of the SEM system, then equations (1) and (2) are estimated on a smaller set of observations (those where Y1 and Y2 are observed) than Equation (3)?
Best regards,
Ruediger
I would like to estimate the following equation:
Code:
Y = A + B*X1 + C*X2 + E
- X1 may be reversely caused by Y. On its own, I would solve this problem by instrumenting X1 with instrument Z1, which is exogenous to Y:
Code:
--ivreg2 Y X2 (X1 = Z1)--
- Whether Y is observed, may also depend on X1, i.e. I have a possible selection problem. On its own, I would solve this problem by first estimating a Heckman Probit model, regressing I(Y!=.) on X1, X2, and Z0, where Z0 should not influence the value of Y:
Code:
--heckman Y X1 X2, select(Z0 X1 X2)--
Put differently, I have two outcomes Y1 and Y2, where Y1 may amongst others depend on Y2, so an alternative to the above IV could be a Simultaneous Equations Model of the form:
Code:
- Y1 = A1 + B1*Xb + C1*Xc + D1*Y2 + EPS1
- Y2 = A2 + B2*Xb + C2*Xc + D2*Z2 + EPS2
Best regards,
Ruediger
Comment