The Statistics and Data Science (SDS) group’s particular emphasis is on the strategic growth area of statistical methods and models for data science, with contributions in the field of experimental design, high-dimensional statistics, inference under censoring and missing data, regression analysis beyond the mean, applied Bayesian inference, statistics of networks and machine learning for complex data situations.
The areas of applications of data analytics investigated by the group include astrophysics, biology and genomics, financial econometrics, health sciences, materials science, wellbeing and UK intelligence, with funding and end user problems coming from industry (TWI, for analyses of pipeline corrosion), the EU (COSTNET project, on network science), ONS (the Office of National Statistics, for analyses of wellbeing data) and overseas collaborators from China, Iran and Italy. There is also particular strength within the group in quantile/expectile regression and its application in bioscience, finance and other areas.
Specific research expertise
- High-dimensional Bayesian Learning (D. Chakrabarty)
- Learning in the Absence of Training Data (D. Chakrabarty)
- Applications of Statistics in Astronomy, Materials Science, etc. using MCMC-based inference (D. Chakrabarty, C. Spire, K. Yu)
- Random Geometric Graphs & Networks (D. Chakrabarty, B. Parker)
- Design of Experiments for Network Science (B. Parker)
- Algorithms for Experimental Design (B. Parker)
- Bayesian regression beyond the mean (K. Yu)
- Weibull analysis for lifetime data analysis (K. Yu)
- Quantile regression for big data (K. Yu)
- Machine learning methods and application (K. Yu, B. Parker)
- Nonparametric smoothing (K. Yu)
- Advanced regression analysis of carbon emissions (K. Yu)
- Applications of Statistics in Health Science, Biology and Genomics (K. Yu)
For more detailed descriptions of research and list of individual publications, please follow the links to the web pages of individual group members.
Major external collaborators
- Prof. Ayan Basu, Indian Statistical Institute, Kolkata, India.
- Dr Rong Jiang, Donghu University, Shanghai, China.
- Prof. Rahim Alhamzawi, University of Al-Qadisiyah, Iraq.
- Prof. Ernst Wit, Universita’ Svizzera Italiana, Switzerland.
Externally funded projects
- UK Intelligence Community Fellowship (RAEng, C. Spire)
- New regression models for the analysis of wellbeing and income distribution (UK Ofﬁce for National Statistics (ONS), 2019-2020, K. Yu)
- High-dimensional and spatial data analysis by Bayesian inference (EPSRC iCase Studentship, 2019-2022, K. Yu).
- Industrial funding to work with TWI’s Asset Life Cycle Management team (2014-2021, K. Yu)
- EU COST Action CA15109 (COSTNET) on Statistical Network Science (V. Vinciotti, B. Parker)
- NIHR Research Methods Opportunity Funding Scheme Application (Ref: NIHR-RMOFS-2013-03-09), (2013-2015, K. Yu (Brunel), J. Lord (Southampton), F. Becker (Oxford))
- News analytics toolkit and stochastic Programming Solution Algorithms (Optirisk Systems, 2012-2013, K. Yu)
- Mathematical Method for Reliability (London Mathematics Society, 2014-2017, K. Yu (Brunel), F. Coolen (Durham)).
- Micro-quantitative and macro-qualitative gene network models from ChIP-sequencing and microarray data (BBSRC, 2011-2013, V. Vinciotti)
- Application of statistical inference for demand modelling and forecasting (EPSRC Industrial Mathematics KTN with Industry Partner: 5one, 2011-2012, K. Yu).
- Optimal allocation of healthcare (NIHR, 2010-2011, K. Yu)
- News data analysis (KTP with Industry Partner: RavenPack/Xenomorph, 2009-2010, K. Yu)