Project description:
A physical process that evolves overtime, such as the fluctuation of an asset price in the financial markets or a stochastic chemical reaction, are examples of a probabilistic dynamical system.

Assimilating data into very high-dimensional probabilistic dynamical models and then simulating the assimilated model forward to make forecasts is the major computational challenge to be addressed in this project. (An exemplar application being Weather forecasting.) However, a direct application of existing data assimilation techniques for time-series models like Particle Filters will not be fruitful as these breakdown as the dimension of the probabilistic dynamical model being studied grows.

This project aims to address this problem of designing new efficient numerical methods for data assimilation in high dimensions by drawing on recent advances in Monte Carlo based Bayesian inference, e.g. Particle Markov Chain Monte Carlo

For further information contact Dr Sumeetpal S Singh: http://sigproc.eng.cam.ac.uk/Main/SSS40)

Applicants should have or expect to obtain by the start date at least a high 2.1 degree, with a strong preference for a 1st class honours, in Statistics or Information Engineering or Machine Learning and this should be preferably a Masters degree. A strong Mathematical background is essential.

This EPSRC funded studentship is available for Home and EU students. Home students and certain EU students will receive a full studentship including fees and Maintenance. EU students will receive a fees only award. Details on eligibility can be found of EPSRC Web site: https://www.epsrc.ac.uk/skills/students/help/eligibility/. Overseas students are not eligible and should not apply.

Applications should be made on-line via the Cambridge Graduate Admissions Office before the deadline: http://www.admin.cam.ac.uk/students/gradadmissions/prospec/apply/ with Dr Sumeetpal S. Singh identified as the potential supervisor

The University values diversity and is committed to equality of opportunity.