Please bring the following to the attention of any suitable people in
your department! (Second announcement.)
PHD STUDENTSHIP IN STATISTICS AT THE OPEN UNIVERSITY
Applications are invited for a full-time EPSRC-funded PhD studentship in
Statistics commencing 1 October, 2009. The studentship will include
University fees and a stipend.
The studentship will be based at the Open University's central campus in
Milton Keynes, where there is a thriving postgraduate student community.
The Statistics Group at the Open University provides a lively and
stimulating environment for Statistics research with active researchers
working in a variety of fields of statistics. For more information about
the Statistics Group see http://statistics.open.ac.uk/
The available project is:
Distributions Constructed With Certain Asymmetry and Gradient Asymmetry
Properties: Approaches, Development and Inference
(details below)
Supervisor: Professor Chris Jones
Applicants should have a first or upper second class honours degree or a
recognized postgraduate qualification containing a substantial element
of Statistics. The usual EPSRC residence requirements apply: see
http://www.epsrc.ac.uk/PostgraduateTraining/StudentEligibility.htm
For the Open University's Research Degrees Prospectus and an application
form, see http://www.open.ac.uk/research/research-degrees/. Please note
that where the standard application form asks for a research proposal,
indicate the project above and describe why you are interested in it.
Informal enquiries may be made by email to Chris Jones (email:
[log in to unmask]). Please note that I will be away quite a bit in May
and so may not respond very speedily.
Closing date for applications: Tuesday, 26 May, 2009.
Outline of project:
This project aims to build on aspects of the paper "Asymmetry and
Gradient Asymmetry Functions: Density-Based Skewness and Kurtosis" by F
Critchley and MC Jones, Scandinavian Journal of Statistics, 35, 415-437,
2008. In particular, by understanding more about the typical shapes and
properties of these "skewness and kurtosis functions" for a wide
range of distributions, it is intended to turn an important question
round in the following way: instead of asking what are the skewness and
kurtosis properties of a given distribution, we ask, given reasonable
simply-parametrised skewness and/or kurtosis functions, what distributions
arise? "Transformation of scale" densities, i.e. ones of the form f(t(x))
[as opposed to transformation of random variables resulting in densities
of the form t'(x)f(t(x))], would appear to have a particular role here,
and the supervisor has already worked on "generalised Schlomilch
transformations" to some effect.
Despite the seemingly arcane terminology, the basic ideas above are all
quite simple, and that is part of their attraction. The student can
develop the project in any of a number of directions, but one would
envisage a mix of foundational and theoretical considerations,
methodological development, and statistical inferential issues, the last
two arising from important questions about the practical worth of this
kind of theoretical development.
|