|
PhD Studentships (EPSRC - CASE)
Department of Probability and Statistics
University of Sheffield
Applications are invited for three EPSRC-funded CASE (Collaborative Awards in
Science and Engineering) PhD studentships in the Department of Probability
and Statistics, University of Sheffield.
Each studentship will be for three years and will include full fees and a
maintenance grant of at least 11,700 pounds per year (under review),
including a contribution of 3,000 pounds per year from the collaborating
body. It is intended that the studentships will start on 23 September 2002
but they could, by arrangement, be delayed until 10 February 2003.
A brief description of each project is given below; further details
(including telephone numbers and email addresses for informal enquiries) are
available on the web at the locations shown.
In addition, one or more other studentships may be available for other
projects; information on possible topics is available on the web at
http://www.shef.ac.uk/~pas/PhD/. Informal enquiries may be made to the PhD
Admissions Tutor, Dr Paul Blackwell ([log in to unmask]; tel.
0114-2223719).
Other information about the department, its staff and their research is at
http://www.shef.ac.uk/~pas/
PROJECT OUTLINES
Probabilistic Relational Models for Gene Expression
Supervisor: Dr. Nick Fieller
Further details: http://www.shef.ac.uk/nickfieller
Collaborating body: AstraZeneca
This CASE studentship involves working with both the EST Statistical Sciences
Department and the Bioinformatics Group of AstraZeneca at Alderley Park,
Macclesfield. The overall aim of the work is to develop tools for analyzing
micro array data. A distinctive feature of such data is that they are very
high dimensional but with relatively few observations. Standard tools of
multivariate analysis (cluster analysis, principal component analysis and
other dimensionality reduction methods) are only partially successful in such
studies. This project will investigate incorporating additional information
in the analyses by application and extension of Bayesian techniques, Bayesian
Network Analysis in particular. This studentship should be attractive to
those with a statistical and/or mathematical background who want to develop
their interest in the direction of bioinformatics.
Statistical Methods for Estimating Former Sea Levels
Supervisor: Professor Clive Anderson
Further details: http://www.shef.ac.uk/~st1cwa
Collaborating body: English Heritage
Past levels of the sea and the changing extent of past wetlands are of much
current environmental and archaeological interest, both in relation to
climate change and also for the light they throw on the spatial and temporal
relationships between past humans and their environment. These questions
have been at the forefront of two recent research projects in the Humber
Wetlands aimed at furthering understanding of the past environment and
archaeology of the area. The work has demonstrated the intimate link between
past sea levels and the mapping of ancient wetland environments, but has
failed to produce robust results fully consistent with archaeological
evidence. Current reconstructions, for example, suggest that some
archaeological sites were under water at times when they are known to have
been occupied. This project will develop statistical tools for estimating sea
levels and wetland extent, using data collected in both the Humber Wetlands
Project and the Land-Ocean Interaction Study (LOIS). In particular it will
quantify uncertainty in palaeoenvironmental reconstruction and palaeotidal
modelling, and will build methodology to integrate model results with
relevant archaeological evidence.
Statistical Modelling for Robust and Flexible Chronology Building
Supervisors: Dr Caitlin Buck, Dr Paul Blackwell
Further details: http://www.shef.ac.uk/st1ceb
Collaborating body: English Heritage
Chronology building plays a vital role in archaeology, geology,
palaeo-environmental studies and palaeo-climatology. In all these
disciplines, methods such as radiocarbon dating are now routinely used to aid
in chronology construction. The current practice of statisticians working
with such data is to assume that material suitable for dating was deposited
between events of unknown calendar date with a uniform deposition rate. It is
clear, however, that for many real projects this assumption is unlikely to
hold. As a result, researchers who use these models would like to understand
more about how this will affect the inferences they make, and be offered some
alternatives for more realistic and robust modelling of deposition processes.
Staff at English Heritage, the CASE sponsor for this studentship, routinely
use Bayesian statistical methods to provide the core of their interpretative
process, but have no statisticians on their team to aid with model
development. The student working on this project will thus need a background
in statistics. They will devise alternative models for the deposition of
datable material and will offer practical assistance with choosing from the
models available. In so doing, they could make an enormous impact on the
day-to-day work of both the industrial sponsor and other archaeologists (and,
potentially, geologists, palaeo-environmentalists and climatologists) engaged
in a wide variety of research around the world.
ELIGIBILITY
Candidates will be expected to obtain at least a II(i) honours degree (or the
equivalent) involving a substantial proportion of mathematics and statistics,
or to already hold such a degree.
Current EPSRC guidelines restrict the maintenance part of the studentship to
applicant with "a relevant connection with the United Kingdom", essentially
to those normally resident in the UK (not resident mainly for educational
purposes). Candidates with a relevant connection to another EU country may
apply for a "fees-only" award.
APPLICATION PROCEDURE
The application procedure is described in the further particulars for each
project. Applications should be made as soon as possible, preferably no later
than 24 May 2002; the Department reserves the right to make an award before
that date.
|
