[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: discrete factor approximations in a simultaneous equation model
I would like to estimate the equation:
y_it = X_it*alpha + Z_it*beta + e_1it,
where Z is endogenous and described by:
Z_it = X_it*gamma + e_2it.
The error structure is given by:
e_1it = v_i + u_1it
e_2it = rho_i*v_i + u_2it,
where v is the common factor specified as the source of correlation between Z and e_1it, and rho_i is a factor-loading parameter to be estimated.
Is Stata capable of estimating this model using a semiparametric ML random-effects estimator like that found in Mroz (J of Econometrics 1999), and if so, has anyone written such a program? I think newer multilevel/multiprocess programs like aML (applied Maximum Likelihood) have canned routines for such estimation, but I am hoping to stick with Stata.
Thanks in advance for any help,
Beck A. Taylor, Ph.D.
* For searches and help try: