Hello.
I'm running a mixture model with a strong assumption of being 80% of people in group 1 and 20% of people in group 2. There are no other parameters I can make any assumptions. Because group 1 is a reference group and some parameters for group 2 are estimated relative to group 1, I need to fix group 1 as group 1. However, because of label switching problem, 80% of people are sometimes assigned in group 1, but sometimes in group 2. Suppose I replicate the same analysis with 10 simulated data. From 7 data, pi[1] (pi is a mixing proportion) is about .80, but from other 3 data, pi[1] is about .20, and pi[2] is about .80. What I thought to resolve this problem is to add an inequality constraint such that pi[1] > pi[2], but it seems this kind of inequality is not working in Openbugs. Is there any suggestion to resolve this issue?
Thanks!
-------------------------------------------------------------------
This list is for BUGS users to support each other, ask questions or share ideas.
To mail the BUGS list, mail to [log in to unmask]
Before mailing, please check the archive at www.jiscmail.ac.uk/lists/bugs.html
Please do not mail attachments to the list.
To leave the BUGS list, send LEAVE BUGS to [log in to unmask]
If this fails, mail [log in to unmask], NOT the whole list
|