Sounds like a compositional data anaysis problem. The classic book is by John Aitchison, Compositional Data Analysis, 2003 (reprint).
Marketers often call this constant sum data. Most models for it have issues from what I recall, but I don't know if any insights from the CDA literature have been applied.
It is on my list of things to do to see how this works for preference data of the type you have so if you are interested, let me know.
-----Original Message-----
From: "Data Analytics Corp." <[email protected]>
To: "Stata Listserve" <[email protected]>
Sent: 6/13/2008 8:45 PM
Subject: st: Estimating a choice model with allocations
Hi,
I have a project coming up that has a discrete choice component.
Doctors will be asked to allocate their next 100 patients to a set of
medications in a choice set. Several choice sets will be shown as part
of an experimental design. If the doctors are asked to choose just one
medication, the estimation could be done with clogit. If they are asked
to rank the medications in a choice set, then rologit could be used.
But what about my case where they have to allocate patients and the
allocations sum to 100? The allocation will, of course, rank the
medications, but rologit seems to require numbers such as 1, 2, 3, ...,
etc. I'll have 10, 20 31, 5, etc. that will sum to 100. How can I
estimate a choice model from this? Any suggestions?
Thanks,
Walt Paczkowski
--
________________________
Walter R. Paczkowski, Ph.D.
Data Analytics Corp.
44 Hamilton Lane
Plainsboro, NJ 08536
________________________
(V) 609-936-8999
(F) 609-936-3733
[email protected]
www.dataanalyticscorp.com
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
