Hi - I wonder if anyone can advise on whether it is possible to fit random
effects Markov models for smoking uptake/smoking cessation probabilities in
BUGS? (or in any other reasonably accessible software for that matter) We
are analysing repeated measurements (say binary: smoker/non-smoker) made in
a cohort of adolescents. We are interested in modelling transition
probabilities, i.e. whether someone starts smoking at a particular time, or
whether they quit given that they were a smoker last time they answered a
question.
The problem is that BUGS will not allow an outcome variable to be defined
as a logical node expressed in terms of previous outcome values, which
seems to be necessary in order to incorporate the Markov structure. When we
try to code this we run into the problem that a data point cannot also be a
node. Is this an insurmountable problem, or are there workarounds? Any
guidance, including guidance on how to layout the data for input into BUGS,
would be gratefully received.
Thanks
Jonathan Sterne
----------------------
Department of Social Medicine
University of Bristol, UK
Address until December 2003:
Clinical Epidemiology and Biostatistics Unit
Murdoch Children's Research Institute
Royal Children's Hospital
Parkville, Vic. 3052. AUSTRALIA
Tel. 9345 6118
All numbers area code 03 or international prefix +61 3
E-mail [log in to unmask]
web www.epi.bris.ac.uk
-------------------------------------------------------------------
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
|