Similarities among scenarios due to the respondent

[Mahdi Samadzad]

Dear professor Bierlaire,

A SP MNL model is specified as U_in = V_in + ɛ_in + η_n where i counts the choice and n counts the respondent. Now how do we account for η in the biogeme? The specification seems to be exactly the same for RP and SP models.

Thank you in advance,

Mahdi.

[Bierlaire Michel]

If you want to deal with serial correlation, you need to declare the SP data as panel. See https://github.com/michelbierlaire/biogeme/blob/working/examples/swissmetro/b12panel.py

> --
> You received this message because you are subscribed to the Google Groups "Biogeme" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to biogeme+u...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/biogeme/b01def0b-ed9c-4386-b2fc-48a428dd8418n%40googlegroups.com.

Michel Bierlaire
Transport and Mobility Laboratory
School of Architecture, Civil and Environmental Engineering
EPFL - Ecole Polytechnique Fédérale de Lausanne
http://transp-or.epfl.ch
http://people.epfl.ch/michel.bierlaire