Print

Print


Dear Statisticians,

Could you please make your students aware of the following PhD opportunity.

Thanks,

Dr Ian Vernon

Durham University


iCASE PhD Studentship in Bayesian Statistics at Durham University, UK.

The Statistics and Probability group at Durham University is offering a well funded iCASE PhD position to study Bayesian uncertainty analysis and decision support for complex models of physical systems. The PhD student will learn the powerful and widely applicable techniques of Bayesian emulation and history matching, techniques that have been successfully applied to the areas of Cosmology, Epidemiology, Systems Biology, Engineering, Environmental Science and Geology. The student will first develop this methodology for application to combinations of complex stochastic and deterministic models. The second half of the project will involve the use of Bayesian decision theory to provide decision support for the control or management of the complex physical system. This is a 4 year project starting in October 2016, supervised by Dr Ian Vernon and Prof Michael Goldstein. This project naturally extends work previously developed by Dr Vernon and Prof Goldstein that was awarded the Mitchell Prize: the top worldwide research prize for applied Bayesian statistics. The developed techniques will be widely applicable across many scientific disciplines, and here they will be tested and applied to expensive geological and reservoir models from the oil industry. 

This is a well funded EPSRC iCASE PhD position. The annual stipend is £17,500 (tax free), which includes the standard EPSRC stipend of £14,296 plus an enhancement provided by an industrial partner, Roxar. There is also plenty of funding for travel to international conferences and for computer equipment, including both a high-end laptop and tablet. 

Roxar, now part of Emerson Process Management, is a mathematical solutions company that services the oil industry. We have a long standing consultancy agreement with Roxar involving the application of Bayesian uncertainty analysis, for which this iCASE PhD studentship provides a natural extension. The student would be expected to liaise closely with contacts at Roxar, and visit their Oxford office for 2 weeks each year and for 2 months in the 3rd year. 

The student will join a substantial team of PhDs, postdocs and staff in the Durham University statistics and probability group conducting research on Bayesian uncertainty analysis of complex models. We have strong collaborations with many high profile research groups across several scientific disciplines including Cosmology, Epidemiology, Systems Biology, Engineering, Environmental Science and Geology. Dr Vernon and Prof Goldstein have also recently been awarded a 5 year BG project in conjunction with UNICAMP, Brazil, which will fund several PhD and postdoc appointments to work on the Bayesian uncertainty analysis of oil reservoirs. The iCASE PhD student will have the opportunity to work closely with this BG team based in Durham, and to travel to Brazil on one or more occasions to further such collaborations.

The candidate would be expected to have at least a 2:1 degree in Statistics, Mathematics or a related subject with a strong quantitative component. No knowledge of oil reservoirs or geology is required. Some knowledge of R would be helpful, but not essential. Funding will be available to cover tuition fees for UK/EU students.


Interested applicants should in the first instance informally apply by emailing a CV and cover letter to [log in to unmask] 

Successful applicants will then be asked to apply formally via https://www.dur.ac.uk/postgraduate/apply/ 

For any informal enquiries, please contact:

Dr Ian Vernon

Statistics and Probability Group

Department of Mathematical Sciences

Durham University, UK

Email: [log in to unmask]

Tel: +44 (0) 191 334 3068

Durham University is an Equal Opportunities employer. See https://www.dur.ac.uk/diversity.equality/

You may leave the list at any time by sending the command

SIGNOFF allstat

to [log in to unmask], leaving the subject line blank.