Statistical Science

The Statistical Science CDT is a four-year DPhil research programme in the theory, methods and applications of next-generation statistical science for 21st century data-intensive environments and large-scale models.  

Graduate destinations

Academia and industry are struggling to find appropriately trained statistical researchers in the key OxWaSP areas of statistical methods and computation for complex data analysis.

The Oxford‐Warwick Centre is dedicated to providing the organisation, environment and personnel needed to develop the future industrial and academic individuals doing world-leading research in statistics for modern day science, engineering and commerce, all exemplified by ‘big data’.

The programme provides structured training and research experience in the first year, followed by a three-year research project leading to a DPhil. It is the Oxford component of OxWaSP (the Oxford Warwick Statistics Programme), an EPSRC and MRC Centre for Doctoral Training in Next-Generational Statistical Science. 

In the first year you will receive training for research in five key areas:

Statistical inference in complex models

The new demands of scientific research and the availability of vast data sets have required statisticians to revisit and reformulate the foundations of their discipline so that theory and methods become scalable to modern data.

Multivariate stochastic processes

A substantial number of inferential environments evolve dynamically in time or space, or both, often under stochastic control. A wide range of applied probabilistic and statistical methods are currently being developed to address these needs.

Bayesian analyses for complex structural information

The recent surge in Bayesian methodologies merges the now well-understood tools of probabilistic reasoning with stochastic computational and statistical inference. Current research frontiers further develop this relationship to apply to an ever increasing domain of application where essential contextual structural information can be properly coded as part of an extensive data-analysis exercise.

Machine learning and probabilistic graphical models

Over recent decades a mutual understanding of the rich symbioses between statistics and machine learning methodologies has developed and researchers have now begun to exploit these relationships. One of the key areas of such exchange is in probabilistic graphical modelling.

Stochastic computation for intractable inference

Many recent advances in statistical modelling have only been made possible by the dramatic progress in techniques which admit the fast analysis of probabilistic and statistical models. These methods are being increasingly customized to the needs of different model classes.

Pattern of teaching, learning and supervision

The first two terms consist of a series of two-week modules. Modules start with two days of lectures. Over the subsequent five days you read some of the original literature and write a report. Industrial and academic speakers visit Oxford for informal lunch sessions mid-module, and you will have the opportunity to invite speakers.

At the end of each module you travel to Warwick for a mini-symposium on the theme of the module. The rest of the first year consists of two ten-week research projects. Towards the end of the first year you choose a supervisor for your main DPhil project in Oxford and carry out this research in years two to four. There will be formal assessments of your progress at around 18 and 36 months into the degree. These assessments involve the submission of written work and oral examination.

The final thesis is normally submitted for examination during the fourth year and is followed by the viva examination.

Where appropriate for the research, student projects will be run jointly with the department’s leading industrial partners and you will have the chance to undertake a placement in data-intensive statistics with some of the strongest statistics groups in the USA, Europe and Asia.

Applicants are normally expected to be predicted or have achieved a first-class or strong upper second-class undergraduate degree with honours (or equivalent international qualifications), as a minimum, in an appropriate subject. You will need a good background in relevant aspects of mathematics and statistics. Success in a degree with a high content of machine learning or a mathematical science (such as physics) may be acceptable.

For applicants with a degree from the USA, the minimum GPA sought is 3.6 out of 4.0.

However, entrance is very competitive and most successful applicants have a high first-class degree of the equivalent.

If you hold non-UK qualifications and wish to check how your qualifications match these requirements, you can contact the National Recognition Information Centre for the United Kingdom (UK NARIC).

No Graduate Record Examination (GRE) or GMAT scores are sought.

• Official transcript(s)
• CV/résumé
• Statement of purpose/personal statement:Up to two pages
• References/letters of recommendation:Three overall, generally academic

ENGLISH LANGUAGE REQUIREMENTS

Higher level

• Global Education
• Hill Foundation Scholarships