Dear Colleagues,
I wonder if anyone might be able to help me. I remember reading an
article that was published, in one of the Royal Statistical Society's
journals I think, around 20 years ago, which I would like to refer to
in a paper I am now drafting. The authors were James and Stein, but
I don't remember the initials. The paper argued that, presumably
within the Bayesian paradigm, the onus is on the researcher or data
analyst to take into account *all* possible prior information,
however seemingly irrelevant. There was a quote which was something
to do with hogs in Taiwan - I think the reference was to either their
price or their weight - as an example of a piece of information that
would have to be taken into account in forming inferences. From my
recollection the authors concluded that it was necessary to take into
account all prior information of this kind, but it was highly
paradoxical that this was so.
I have hand searched for this article in my stock of RSS journal
issues, without success. I would be extremely grateful if anyone
could identify this paper for me and email the full reference to me.
Please reply to me and not to the list - and please don't take this
request as a prompt to reiterate the heated arguments between
proponents of Bayesian and frequentist inference!
Many thanks.
Robert Newcombe.
..........................................
Robert G. Newcombe, PhD, CStat, Hon MFPHM
Senior Lecturer in Medical Statistics
University of Wales College of Medicine
Heath Park
Cardiff CF14 4XN, UK.
Phone 029 2074 2329 or 2311
Fax 029 2074 3664
Email [log in to unmask]
|