I need to read a data set that has multiple observations per person (like scanner panel data). I want to use optim and compute the objective function person by person where each person has multiple observations (as in latent class models a la Kamakura and Russell (JMR, 1989))? I tried to use the (for) function but could not get it to do "Do loops" like in GAUSS or FORTRAN. Even if you have components of the program I am looking for, it could help. If so, would you send it to me at fusungonul at hotmail.com? Thank you. ----- Thank you, Fusun -- View this message in context: http://r.789695.n4.nabble.com/Unobserved-heterogeneity-tp2324520p2324520.html Sent from the R help mailing list archive at Nabble.com.