[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Instrumental variable est
I have a problem with the instrumental variable estimation. It turns out that there are two endogenous variables in a system, but one of then is a dummy variable, and itís related with the probability of certain event. In normal circumstances, I would estimate this dummy variable by using a probit regression, but because of its endogenous condition, I thought it would be better to estimate the entire system by two-stage least squares. The problem here is just that Iím not sure that using the instrumental variable estimation would be the better solution, because there is a dummy variable involved, and I donít know a command that could merge the probit estimation of one equation and the least square estimation of the another equation of the system.
Has anybody a recommendation?
Obtenga su E-mail GRATUITO en http://economistasuanl.zzn.com
Para obtener su propio servicio de correo electrůnico basado en la Multimalla, dirŪjase a http://www.zzn.com
* For searches and help try: