Contact Details
| Email Address (Work) | simon.richard.white >at< mrc-bsu.cam.ac.uk |
| Email Address (Personal) | simon.richard.white >at< 3-14.co.uk |
| Telephone (Direct): | +44 (0) 1223 7 [67408] |
| Office | F40 |
| Address |
MRC Biostatistics Unit Institute of Public Health University Forvie Site Robinson Way Cambridge CB2 0SR UK |
Brief Biography
Currently hold a Career Development Fellowship at the MRC Biostatistics Unit, part of the Institute of Public Health in Cambridge, working with Professor Sheila Bird on a variety of public health issues.
I completed my PhD in statistics during 2009 at the University of Nottingham's School of Mathematical Sciences, supervised by Professor Philip O'Neill. My research considered parameter inference using Markov Chain Monte Carlo for the SIR stochastic epidemic model conditioned on the final outcome of the epidemic.
From 2001-2005 I completed a taught undergraduate masters (MMath) at the University of Nottingham's School of Mathematical Sciences.
Research Interests
I am interested in a wide variety of statistical methods and applications, specifically Bayesian statistics and associated methods.
In brief, they include:
- Statistics applied to public health issues, in particular related to blood-borne infectious diseases.
- Promoting and supporting proper statistical methodology in research.
- Bridging innovative statistical research and practical applications to the non-statistical community.
- Application of Bayesian methodology, in particular Markov chain Monte Carlo.
- Developing the theory of Approximate/Exact Bayesian Computation (ABC/EBC) and applications to stochastic models.
- Stochastic epidemic modelling and inference, particularly on partially observed epidemics.
- High performance statistical computing, parallel and distributed applications (OpenMP/MPI/GPU).
Epidemic Modelling
Analysing epidemics that have occurred is difficult given limited information about the disease and other factors of the situation. We can approach the problem by considering final size data, generally commonly reported, as well as extending to include information on social groupings and interactions.
Using this limited information, and making restrictive assumptions about the mechanism of infectivity and spread of the disease, we make inference about the parameters of the model. There are several limitations to this method, not least its reliance on specified model assumptions and simplifications. Also, the technique used does not allow for any temporal information to be incorporated - a significant area of current research.