Dear Colleague,

We would be most grateful if the following advertisement could
be drawn to the attention of suitable candidates.

Many thanks,
Dr. Philip D. O'Neill
Prof. Gareth O. Roberts

------------------------------------------------------------


      UNIVERSITY OF NOTTINGHAM and UNIVERSITY OF LANCASTER 
      ----------------------------------------------------

       PhD STUDENTSHIP and POSTDOCTORAL RESEARCH ASSISTANT 
       --------------------------------------------------

Applications are invited for the two above-named posts in connection 
with the following EPSRC-funded project.  Both posts are funded for 3
years, and are available from 1st Oct 1999 (although there is some
flexibility about the exact start date.)

(Note: the usual EPSRC eligibility requirements for studentships do 
NOT apply, and so applicants can be from outside the UK and EC.) 


== MARKOV CHAIN MONTE CARLO METHODS FOR STOCHASTIC EPIDEMIC MODELS ==

Background: When analysing data from infectious disease outbreaks 
it is often desirable to use stochastic models which describe 
the underlying infection process.  However performing statistical 
inference for such models is complicated by a number of factors, 
such as the high level of dependency in the data and and fact 
that data is invariably only partial.

Recently, the use of Markov chain Monte Carlo (MCMC) methods has 
been explored for such problems.  Such methods generally allow for
flexible modelling assumptions and result in a more powerful 
analysis than has previously been possible.  However, for a number
of reasons standard MCMC methods can encounter severe convergence 
problems for certain modelling situations.

The Project: The main aim of the project is to develop and apply 
efficient MCMC algorithms for Bayesian inference for stochastic 
epidemic models.  Three main kinds of modelling situation will be
considered, namely epidemics in large single populations; epidemics
among households; and epidemics in populations with social structure.
The project brings together a number of topics such as: developing 
new MCMC algorithms based on recent theoretical advances; performing 
Bayesian analyses for infectious disease data; developing and 
analysing stochastic epidemic models.  In particular, there is a blend
of applied probability, stochastic processes, statistical inference, 
and computation methods.


Postgraduate Research Studentship 
==================================

The Research Studentship, expected to result in a PhD, is based in
the Division of Statistics in the School of Mathematical Sciences 
at the University of Nottingham.  The Division, which achieved a 
Grade 4 rating in the last Research Assessment Exercise, is 
currently enjoying a period of expansion, with four recent staff 
appointments creating a total of 9 academic staff.  There are also 
currently 8 research students and research staff.  The Student would 
join an internationally-recognized research group in Stochastic 
Epidemic Models.  The Studentship project will be mostly (but not 
exclusively) concerned with models for the spread of epidemics among 
groups of households, and applying the models to real data.  The 
funding includes tuition fees, funding for training, and a standard 
EPSRC maintenance grant.

Applicants for the Studentship should have, or will shortly obtain, a 
first or good upper second degree or Masters in Mathematics or 
Statistics.


Postdoctoral Research Assistant
================================

The Postdoctoral Research Assistant would be based in the Statistics
Group in the Department of Mathematics and Statistics at the University 
of Lancaster.  The Statistics Group includes 12 academic staff and 
25 research students and research associates, and achieved Grade 5 
rating in each of the last two Research Assessment Exercises.
The RA would be concerned with the development and implementation of
state-of-the-art MCMC methods, particularly with a view to application 
to single-outbreak epidemic models and network epidemic models.  The
salary commences at point 6 of the RA/1A scale (currently GBP 17,570).

Applicants for the Postdoctoral RA should have, or will
shortly obtain, a PhD in Statistics or Applied Probability.


Project Team
============

The two principal investigators are Dr. Philip O'Neill (Nottingham - 
currently at Liverpool) and Prof. Gareth Roberts (Lancaster).  In 
addition, Prof. Niels Becker (ANU, Australia) is acting in an advisory 
capacity.  The project team will have regular meetings in both 
Nottingham and Lancaster, and it is anticipated that there will also 
be opportunities to visit Prof. Becker in Canberra.  Funding is also 
available for conference attendance.


Applications
============

Applicants should send a CV (two copies) by post or email to the
appropriate address below. Informal enquiries are also welcome.  
The closing date for applications is July 9th although late 
applications may also be considered.

(Studentship)
Dr. P.D. O'Neill, Division of Statistics and OR, Department of
Mathematical Sciences, University of Liverpool, Liverpool L69 3BX
Email: [log in to unmask]
Tel: 0151-794-4737

(Postdoctoral position)
Prof. G.O. Roberts, Department of Mathematics and Statistics, 
University of Lancaster, Lancaster LA1 4YF
Email: [log in to unmask]
Tel: 01524-593948







%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%