Computational Statistics
We're at the forefront of developments in the efficient analysis of big data sets.
Computational Methods
Computational methods are an essential link connecting data to statistical models and learning. Our group has diverse expertise in producing methods for large or complex datasets, as well as in their performance analysis. Some key interests include:
- Monte Carlo methods
- Bayesian methods and their approximation
- Methods for network-valued or heavy-tailed data
- Emulation of complex statistical models
- Applications in engineering and biology
Doctoral training in cloud computing for big data
Joint with the School of Computing, we run an EPSRC Centre for Doctoral Training in Cloud Computing for Big Data. This provides around five new PhD students each year. Students work on projects involving:
- time-series data
- large spatial data sets
- genomics