I would like to use the predicted values from one regression as explanatory
variables in another regression. The model looks like,
Y=g(Z) (1)
P=f(X,Y) (2)
None of the explanatory variables from equation (1) are included in eq.
(2), only the predicted values for the dependnet variable. The dependent
variable in eq. (2) is not included as an explanatory variable in eq. (1).
Questions:
(i) How do I this as a system in STATA (which I suppose that I have to
since I have "errors in variables") when eq. (1) is linear and eq. (2) is
nonlinear?
