EPSRC-ESRC Collaborative PhD Studentship
at The University of Manchester
Applications are invited for a three-year full-time PhD studentship
in the School of Mathematics and School of Social Sciences at The
University of Manchester. The post is funded by the EPSRC-ERSC
Collaborative Studentship Programme, which offers a maintenance
allowance of 12,300 pounds per annum over three years and will
cover all university fees at EU rate. The project is entitled
"Multilevel modelling with heterogeneous variances” and will be
supervised jointly by Prof Jianxin Pan (School of Mathematics)
and Dr Mark Tranmer (Cathie Marsh Centre for Census and Survey
Research, School of Social Sciences). For more information about
the School and the Centre, see our websites at
http://www.maths.manchester.ac.uk/ and http://www.ccsr.ac.uk/.
A brief description of the project is given at the end of
this email.
The studentship is open to the UK & other EU citizens but an
EU national who has not been resident in the UK for the last
3 years is eligible for a fees award only.
Applicants should be interested in working on social science
applications of statistics and have at least a 2(i) Honours
degree or a recognised postgraduate qualification containing
a substantial element of statistics.
Applications (in the form of a c.v. and covering note) and
enquiries should be directed to Prof Jianxin Pan at
[log in to unmask] (Tel: 0044 161 2755864) or
Dr Mark Tranmer at [log in to unmask]
(Tel: 0044 161 275 4744). An application form can be downloaded
from the School of Mathematics, The University of Manchester at
http://www.maths.man.ac.uk/postgraduate/pgadmission/apply.html
The position is available immediately. The starting date is open
for discussion but we would anticipate this being no later than
February 2007.
The deadline for applications is Thursday 30th November 2006.
Brief Project Description:
Multilevel modelling with heterogeneous variances
Standard multilevel models assume that all individuals in a group
have an equal and identical group effect. For example, all pupils
who attend a particular school are equally affected by a ‘school
effect’ for that school. In practice, some pupils in a school may
be affected in different ways from others. We will investigate this
phenomenon by modelling the variance structures in multilevel models
in terms of regression models. We will also study multilevel modelling
with restricted data. Compared to a standard multilevel dataset that
contains all information on individuals and group membership for a
range of outcomes of interest, what are the consequences of estimating
the models when we have restricted information? For example, we do
not know for all individuals, the group identifiers, we only have
information at group level, or we only have information about some
of the groups. This project aims to develop new statistical
methodologies for multilevel modelling to cope with those problems
and to apply to a variety of social science practices. Real data
analysis and computer simulation will be made.
----------------------------------------------------------------------
Prof Jianxin Pan
Professor of Statistics
School of Mathematics
University of Manchester
PO Box 88, Sackville Street
Manchester M60 1QD
UK
Tel: 0044 161-27-55864
Fax: 0044 161-30-63220
Email: [log in to unmask]
Web page: http://www.ma.man.ac.uk/~jpan/
------------------------------------------------------------------------
|