I beleive that there is a similar procedure proposed for logit in the
Journal of Econometrics. One by Seuyoshi, one by George Borjas.
Maybe together.
jacob
Marzio Galeotti wrote:
> >Dear All,
> >
> >I intend to estimate a logit model on individual data. Among the
> regressors I intend to include a regional explanatory variable, i.e. a
> variable at a higher level of aggregation.
> >
> >I know that a problem arises addressed by Moulton in his paper "An
> Illustration of the Pitfall in Estimating the Effects of Aggregate
> Variables in Micro Units" (REStat, 1990). I also know that in the case of
> linear model a two stage regression procedure will fix the problem.
> >
> >The question is how to go about that problem in nonlinear models such as
> logit ones.
> >
> >Is there anyone familiar with the issue who would be so kind as to suggest
> how operationally correct the problems created by aggregate explanatory
> variables in logit models estimated on individual data.
> >
> >Thanks in advance.
> >
> >Marzio Galeotti
> **************************************
> * Prof. Marzio Galeotti *
> * Dipartimento di Scienze Economiche *
> * Università degli Studi di Bergamo *
> * Piazza Rosate 2, I-24129 Bergamo *
> * Ph. +39-035-277540 *
> * Fax +39-035-249975 *
> * E-mail [log in to unmask] *
> **************************************
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|