On Fri, 7 Aug 2009, Zhenzeng Young wrote: > When i was estimation a simultaneous equation model, a error massage > appeared: You must include a constant in this sort of model. > Could anybody be kindly tell me the reason and how to solve the problem? > Thanks! > PS. system setting: > # set the model up as a system > > system name="all" > equation pf 0 gdpr pcr kaopen > equation pcr 0 pf gdp topen fdi efw > endog pf pcr > end system > > # and estimate it in various ways > estimate "all" method=ols In the script in your email, the "equation" and "endog" lines started with funny non-ASCII symbols. I'm not sure if that's the issue. But your equations contain "0" (= the constant) OK. I tried mocking up your example thus <script> nulldata 100 series pf = normal() series gdp = normal() series gdpr = normal() series pcr = normal() series kaopen = normal() series topen = normal() series fdi = normal() series efw = normal() system name="all" equation pf 0 gdpr pcr kaopen equation pcr 0 pf gdp topen fdi efw endog pf pcr end system estimate "all" method=ols </script> This script ran without any error. What version of gretl are you using? Allin Cottrell _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users