Bayesian Statistics
The School has long had interests in the use of modern computationally intensive Bayesian methods for solving difficult statistical inference problems involving complex variation. In recent years, particular interests have developed in the application areas of bioinformatics and systems biology. There is therefore considerable overlap between this group and the Statistical Bioinformatics group. More generally, our research covers many areas of application including screening methods, econometrics, design of experiments, environmental extremes and non-linear system control. Methodological interests encompass simulation methods for very large sparse linear models, multivariate forecasting, and inference for partially observed stochastic processes. A recent grant enabled purchase of a Beowulf cluster to develop and test effective parallel computing strategies for computationally-intensive Bayesian inference for very large lattice Markov spatio-temporal models.
There is considerable overlap between this group and the Statistical Bioinformatics group.
|
Prof. Richard Boys
|
|
|
Dr Malcolm Farrow
|
|
|
Dr Lee Fawcett
|
|
|
Dr Andrew Golightly
|
|
|
Dr Daniel Henderson
|
|
|
Dr Jian Shi
|
|
|
Dr David Walshaw
|
|
|
Prof. Darren Wilkinson
|
|
|
|
Grants and fellowships
Current
- CaliBayes - BBSRC Bioinformatics and e-science programme II
- ComparaGRID - BBSRC Bioinformatics and e-science programme II
Recent
- Hidden Markov models for DNA sequence segmentation - University Fellowship held by Dr Daniel Henderson
- Bayesian inference for stochastic kinetic genetic networks - BBSRC Bioinformatics initiative II
- Parallel Bayesian Computation - Royal Society
Some key recent publications
M. Aitkin, R.J. Boys, T.J. Chadwick (2005). Bayesian point null hypothesis testing via the posterior likelihood ratio. Statistics and Computing 15, 217 - 230.
M. Farrow (2003). Practical building of subjective covariance structures for large complicated systems. Statistician 52, 553-573.
M. Farrow, M. Goldstein (2005). Trade-off sensitive experimental design: a multicriterion, decision theoretic, Bayes linear approach. Journal of Planning and Inference 42, 553-573.
Fawcett, L., Walshaw, D. (2006) A hierarchical Model for Extreme Wind Speeds, Applied Statistics, in press.
Fawcett, L. and Walshaw, D. (2006) Markov Chain Models for Extreme Wind Speeds, Environmetrics, in press.
M. Goldstein, D.J. Wilkinson (2001). Restricted prior inference for complex uncertainty structures. Annals of Mathematics and Artificial Intelligence 32, 315-334.
A. Golightly, D.J. Wilkinson (2005). Bayesian inference for stochastic kinetic models using a diffusion approximation. Biometrics 61, 781-788.
A. Golightly, D.J. Wilkinson (2006). Bayesian sequential inference for nonlinear multivariate diffusions. Statistics and Computing. (to appear).
J.Q. Shi, R. Murray-Smith, D.M.Titterington (2003). Bayesian regression and classification using mixtures of Gaussian process. International Journal of Adaptive Control and Signal Processing 17, 149-161.
J.Q. Shi, R. Murray-Smith, D.M. Titterington (2005). Hierarchical Gaussian process mixtures for regression. Statistics and Computing 15, 31-41.
E.L. Smith, D. Walshaw (2003). Modelling Bivariate Extremes in a Region. Bayesian Statistics 7, Oxford Science Publications, Oxford, 681-690.
D. J. Wilkinson (2005). Parallel Bayesian Computation, Chapter 16 in E. J. Kontoghiorghes (ed.) Handbook of Parallel Computing and Statistics, Marcel Dekker/CRC Press, 481-512.
D.J. Wilkinson, S.K.H. Yeung (2002). Conditional simulation from highly structured Gaussian systems, with application to blocking-MCMC for the Bayesian analysis of very large linear models. Statistics and Computing 12, 287-300.
D.J. Wilkinson, S.K.H. Yeung (2004). A sparse matrix approach to Bayesian computation in large linear Models. Computational Statistics and Data Analysis 44, 493-516.