Walt, you are probably speaking about Discrete Choice Models based on the McFadden Random Utility Models. These models are not implemented in gretl. You can estimate them in STATA or Nlogit (retail software) R or Biogeme (open source) See http://www.youtube.com/watch?v=F1Zj7W_z8yk how to estimate Discrete Choice Models in Stata and R -- Petr Mariel Departamento de Econometría y Estadística UPV, Facultad de Ciencias Económicas y Empresariales Avda. Lehendakari Aguirre 83 E48015 BILBAO, SPAIN e-mail: Petr.Mariel(a)ehu.es Tel.: +34.94.601.3848 Fax.: +34.94.601.3754 --------- Mensaje reenviado -------- De: Sven Schreiber <svetosch(a)gmx.net&gt; Reply-to: Gretl list <gretl-users(a)lists.wfu.edu&gt; Para: walt(a)dataanalyticscorp.com Cc: Gretl list <gretl-users(a)lists.wfu.edu&gt; Asunto: Re: [Gretl-users] estimating discrete choice models Fecha: Mon, 17 May 2010 14:29:07 +0200 Data Analytics Corp. schrieb: Well if the three alternatives are different from one observation to the next (why are they always three then, though?), this is beyond me, sorry. Probably not doable in gretl (apart from programming it up yourself with gretl's scripting language, of course...). good luck, sven _______________________________________________ Gretl-users mailing list Gretl-users(a)lists.wfu.edu http://lists.wfu.edu/mailman/listinfo/gretl-users