Non-parametric priors for Bayesian inference
Project: Non-parametric priors for Bayesian Inference
Supervisor: Fabrizio Leisen
The use of Bayesian non-parametric (BNP) priors in applied statistical modelling has become increasingly popular in the last few years. From the seminal paper of Ferguson (1973, Annals of Statistics), the Dirichlet Process and its extensions have been increasingly used to address inferential problems in many fields. Examples range from variable selection in genetics to linguistics, psychology, human learning, image segmentation and applications to the neurosciences. The project aims to introduce and/or investigate new Bayesian non-parametric priors that can be used to describe a wide spectrum of phenomena.
Keywords include:
- Dirichlet and Pitman Yor Processes
- Vectors of Dependent random probability measures
- Non-exchangeable species sampling sequences
The School of Mathematics, Statistics and Actuarial Science
The School offers a lively research culture with a thriving group of around 250 taught postgraduates students and almost 100 postgraduate students, postdoctoral researchers and academic staff. Kent is among the best research-intensive universities in the UK. In the 2008 Research Assessment Exercise (RAE), the university was ranked 24th out of 159 institutions for its world-leading research, with six of our subject areas (including statistics) in the top tennationwide. Details of the research undertaken by the School can be found at:
http://www.kent.ac.uk/smsas/postgraduate/phd-applications.html
The funding
This scholarship is partially funded by the FP7 Marie Curie Career Integration Grant “Flexible Bayesian non-parametric priors” held by Fabrizio Leisen. The scholarship is offered at the standard UK Research Councils' rate (currently£13,863) and will cover the tuition fees for a period of 3.5 years subject to satisfactory progress at each review period.
Entry Requirements
All candidates must have an excellent academic track record in Mathematics or Statistics or related disciplines and hold a good Honours degree in a relevant subject or equivalent. A Masters qualification such as an MSc or MMath orresearch experience would be advantageous.
To apply
To apply complete the online application form available by clicking the 'Apply' button below or by visiting: http://www.kent.ac.uk/studying/postgrad/apply/index.html and select School research Scholarship under `How are you intendingto fund your studies?' and mention this research project in the research proposal. Closing date: 31st March 2014.