RE : Modeling of different formulations

From: "GIRARD PASCAL" PASCAL.GIRARD@adm.univ-lyon1.fr Subject: RE : [NMusers] Modeling of different formulations Date: Tue, March 15, 2005 8:32 am Dear Dennis, Why not simply treating the formulation as a covariate and having 2 separate fixed effect parameters THETAs and random effect ETAs, one corresponding to 1st formulation, and the 2nd one to the other formulation? Pascal =========================================Dr Pascal Girard EA 3738 Ciblage Thrapeutique en Oncologie Fac Mdecine Lyon-Sud BP12 69921 OULLINS Cedex France Tel 33 4 78 86 31 53 Fax 33 4 78 86 31 49 e-mail : Pascal.Girard@adm.univ-lyon1.fr