Hello, I am a new user there. I'd like to know the steps in order to add instrument variables to a single exogenous variable. In Gretl it seems that you are able to add as many instruments as you want, but you cannot specify the variable to which the instrument is related. Thank you for your attention.

Am 28.11.2017 um 18:36 schrieb Riccardo (Jack) Lucchetti: > On Tue, 28 Nov 2017, Sofia Gori wrote: > >> Hello, >> >> I am a new user there. I'd like to know the steps in order to add >> instrument variables to a single exogenous variable. >> In Gretl it seems that you are able to add as many instruments as you >> want, >> but you cannot specify the variable to which the instrument is related. >> Thank you for your attention. > > You don't have to. All the instruments act simultaneously on all > regressors. This is not something gretl-specific, this is the way IV > estimation works. Of course Jack is right, but in case the question was meant as to how certain variables are marked as "endogenous" (and thus needing instruments): You announce certain regressors to gretl as exogenous by also including them in the list of instruments. Then by implication, all regressors that are _not_ instruments are treated as endogenous. (A quotation from the TSLS/IV help: "Note that any exogenous regressors should appear in both lists." [independent variables and instruments]) cheers, sven _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users