Next statistics seminar will take place on
18 October at 3:10pm, room JCMB5323.
There will be tea/coffee after the talk.
Speaker: Richard Arnold (Wellington, NZ)
"Interrogation Theory"
Abstract:
The goal of an investigation, scientific or otherwise, is to find the
answers to some specific set of questions about the state of nature.
Decision theory and the theory of experimental design provide methods to
optimise the design of the investigation and the estimation of the
result. However, this theory is normally set in the context of a
particular model of the universe, with its particular parameterisation.
This requires the investigator to specify a priori a coherent utility of
outcomes with respect to that parameterisation. Quite commonly, the
investigator may not be able to do this.
Ideally an investigator would be able to merely pose a set of questions,
define a set of constraints on the data types, acquisition cost and
logistics, and provide a functional to translate the questions into any
particular parameter space. Theory and methodology would then
semi-autonomously drive the interrogation of the state of nature by
optimally selecting one or more relevant models and parameter spaces,
and designing, acquiring and analysing data, in order to best answer the
questions. We present such a theory of interrogation in this talk.
We review the relevant aspects of decision and design theory, and cast
them in a framework where the investigator specifies a utility only at
the level required by the general questions posed by the investigator.
Each model under consideration is then mapped into this space of
possible answers. We then extend this framework to sequential
investigations, where the outcome of each step may affect all aspects of
the decision problem: the models entertained, the utilities, and even
the questions themselves. We illustrate the theory on a range of well
known and lesser known interrogation problems.
This work is joint with Andrew Curtis (School of GeoSciences, University
of Edinburgh)
This seminar is joint with BioSS, and it is a part of Maxwell institute
seminar series.
Website: http://www.maths.ed.ac.uk/events/statistics
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|