[Limdep Nlogit List] Forcing groups in the Latent Class model

Phil James pasj500 at york.ac.uk
Wed Apr 8 23:48:57 EST 2009

Dear All,


I was wondering if it was possible to force Nlogit to segment individuals'
preferences based on their membership of a group.


For example I have a set of choices made by 200 individuals, who can be
classified as early, late and recent participants of a local authority
scheme. Can I identify choice preferences for those who are early
participants, late participants and recent participants without assessing 3
separate models? Forcing three groups using the standard commands pts=3 does
not obviously divide the individuals by group rather divides them based on
their preference structures.


Many thanks



- - - - - - - - - - - - - - - - - - 

Philip James

University of York



YO10 5DD




More information about the Limdep mailing list