Imperial College London
Department of Mathematics
The Statistics section seeks applications from highly motivated individuals for a number of funded PhD opportunities.
The section has wide interests, embodied in the following large research groups: Time Series and Wavelets, Classification and Data Mining, Statistical Methods in Retail Finance, Statistical Genetics and Bioinformatics, and Statistical Methodology.
The following studentships are intended to start in October 2006. For informal discussion, please contact the named supervisor, or Dr Niall Adams ([log in to unmask]) for more details. Closing date for applications is 2 June 2006.
1).
Pattern Discovery and Detection.
Supervisor: Professor David J. Hand ([log in to unmask])
The National Patient Safety Agency collates data on patient safety incidents across the UK, with the aim of detecting common features and
developing preventative strategies. This project will apply theory and methods developed for pattern discovery and anomaly detection to
this problem. This is an EPSRC CASE award. There are many open questions in this area.
2).
Distortion in Multivariate Streaming Data.
Supervisor: Professor David J. Hand ([log in to unmask])
Many real problems involve modelling dynamic sources generating real-time multivariate streaming data which is distorted. This
student will be part of a team developing methods for coping with different kinds of distortion. In particular, this studentship will
focus on cases when some of the data streams may be subject to measurement error, bias, and variance and covariance distortion, with
both random distortion and non-ignorable distortion.
3). Bayesian Methods in Metabonomics
Supervisors: Dr Maria De Iorio (Biostatistics)
Dr Dave Stephens ([log in to unmask])
The modelling and statistical analysis of metabonomics data is one of the most important areas of scientific investigation in the biological
and life sciences. Advanced statistical, mathematical and computation techniques are vitally important in the success of this crucial
technological area. The project will focus on the use of Bayesian statistical inference procedures, and the development of computational
tools that can be used to analyze metabonomics data; specifically, the three types of analysis that will be used will be based on: flexible
models based on wavelets and other basis functions; non-stationary time-series and state-space models; and finite mixture model.
further details at
https://icex.imperial.ac.uk/exchweb/bin/redir.asp?URL=http://www.ma.imperial.ac.uk/showparticularsAdJDFP.php?rowid=12
4). Customer Management Strategies
Supervisor: Professor David J. Hand ([log in to unmask])
Important issues in personal banking are tracking the effectiveness of customer management strategies, choosing between different strategies
as early as possible, and studying the sensitivity of bad debt to outside influences. This project will be in collaboration with a
financial institution.
5). Predictive Models for Consumer Credit Portfolios
Supervisor: Professor David J. Hand ([log in to unmask])
In the personal banking sector it is important to be able to identify debt portfolios which are performing badly, so that appropriate
actions can be taken. The aim of this project is to develop predictive models which can be used to allocate resources efficiently
across a range of consumer debt portfolios. This project will be in collaboration with a financial institution.
6). Applications are invited for a funded position in areas relevant to the research groups described above.
To apply, please obtain an application from
http://www.ma.ic.ac.uk/pgstudies.php
Return the completed application to
Admissions, Registry
Level 3, Sherfield Building
Imperial College London
South Kensington campus
London SW7 2AZ
|
