Dear BUGS users and developers,
I would like to model the clinical presentation of a disease and outcome at the individual level. The parameters are multivariate and categorical! Although individual feature can be treated as bernoulli or binominal, the should not be since they WILL BE correlated with each other and NOT independent. The probabilities of ecah outcome will not be independant of the others. Essentially we will be having a large number of conditional probabilities which will most likely fail in BUGS.
Multinominal is not valid since the individual events are NOT EXCLUSIVE.
What we need is a bernoulli AND/OR binominal counterpart of the multivariate distribution. which can be specified by giving correlated priors similar to specifying a multivariate distribution.
Is there a way of doing it with existing disttibutions. Wont it be a good idea to create such a distribution? Granted there are not too much clinicians learning or using WinBUGS, but BUGS will be very useful for clinical situations.
My second problem related to incorporating the Hardy Weinberg equilibrium into the multinominal. Although the distribution of genotypes in groups of study subjects can be treated as multinominal (three possible outcomes for a bi-allelic locus), the priors are not appropriately specified as Derichlet since the Hardy Weinberg equilibrium constrains the distribution further and we need some way to get around it.
-------------------------------------------------------------------
This list is for discussion of modelling issues and the BUGS software.
For help with crashes and error messages, first mail [log in to unmask]
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
