---------------------------------------
RESEARCH FELLOW IN STATISTICS/FISHERIES
=======================================
University of St Andrews, Scotland
School of Mathematics and Statistics
Closing Date: 17th February 2000 Ref: ML101/SMC1-00
This EC funded project will develop statistical methodology for
improved model selection and modelling of interactions in Generalized
Additive Models (GAMs) and will apply the methods to modelling of
sardine and anchovy egg abundance and spatial distribution off Spain
and Portugal. The post is based in the large and active Research Unit
for Wildlife Population Assessment (RUWPA). The statistical questions
to be addressed are: how should the degrees of freedom in a GAM be
chosen? How should interactions be modelled in a GAM setting? How can
these things be done in a computationally efficient way? This will
build on work already done at St. Andrews (http://www.ruwpa.st-
and.ac.uk/simon/gam.html).The biological questions are: what is our
best abundance estimate for sardine and anchovy eggs (and hence
adults) in the populations of interest, and how are these abundances
changing over time? What are the environmental determinants of egg
abundance/distribution? Answering these questions is important for
management, and for assessing the impact of environmental change.
Earlier work on this is at: http://www.ruwpa.st-
and.ac.uk/projects/fish/mackerel/mack.html. The ideal candidate will
have a good honours degree and preferably a PhD in a quantitative
subject with a statistical component. Experience with some of the
following would be an advantage: GLMs/GAMs; C/C++ ; Splus/R;
fisheries assessment methodology.
The post is initially for 30 months with salary at the appropriate
point on the 1A or 1B scale for research staff, i.e. 16,286- 24,479
pounds sterling per annum. Direct informal enquiries to Dr. Simon
Wood (tel. 01334 463799, e-mail [log in to unmask]).
The successful candidate will join the Research Unit for Wildlife
Population Assessment (RUWPA). RUWPA is a contract-funded research
group in the Statistics Division of the School of Mathematics and
Statistics which provides advice and assistance to a wide variety of
organisations, including government agencies, conservation groups,
research institutes and wildlife managers. It specialises in the
development of new statistical methods and innovative applications of
existing methods to wildlife population assessment problems. The
group currently comprises seven full-time researchers and a half-time
administrator. Two permanent staff and five postgraduate students in
the Statistics Division have active research interests that overlap
substantially with RUWPA's work (see http://www.ruwpa.st-and.ac.uk/).
Members of the group and Statistics Division have expertise in most
aspects of wildlife assessment and sighting survey design. Although
this post is initially for 30 months there is a strong chance of
renewal to work on other RUWPA projects.
RUWPA is the largest research group within the newly created Centre
for Research into Ecological and Environmental Modelling (CREEM).
CREEM was established in early 1999 to promote collaborative research
on ecological and environmental modelling between the Schools of
Mathematical and Computational Sciences, Environmental and
Evolutionary Biology, and Geography and Geosciences.
Application forms and further particulars from Personnel Services,
University of St Andrews, College Gate, North Street, St Andrews,
Fife KY16 9AJ, (tel: +44-1334 462571 (24hrs), by fax +44-1334 462570 or by
e-mail [log in to unmask]). We regret applications cannot be
made by e-mail. The University operates Equal Opportunities and No
Smoking Policies.
--------------------------------------------------------------------
David Borchers
RUWPA, Maths Institute, North Haugh ph: (+44) 1334 463806
University of St Andrews fax:(+44) 1334 463748
Fife, KY16 9SS, Scotland email: [log in to unmask]
web: http://www.ruwpa.st-and.ac.uk
--------------------------------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|