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. Our research covers many areas of application and, in recent years, has focussed increasingly in the application areas of bioinformatics and stochastic systems biology. Therefore there is considerable overlap between the work of this group and that undertaken in the Statistical Bioinformatics and Stochastic Systems Biology group.

Our main areas of interest are:

  • inference for kinetic rate constants in stochastic/deterministic models using partially observed experimental data, particularly using simulators and emulators
  • data integration in bioinformatics
  • inference for dynamic models of population movement in the Neolithic period
  • phylogenetics, particularly the origin of the eukaryotes
  • Bayesian nonparametric functional regression analysis
  • Bayes linear methods
  • multivariate time series and forecasting
  • representation of substantive prior information for covariance matrices
  • spatio-temporal modelling of rainfall data
  • experimental design
  • environmental extremes

Current grants, fellowships and studentships

In addition to those listed in the Statistical Bioinformatics and Stochastic Systems Biology group:

  • Social media analysis for social geography (CI: Prof Darren Wilkinson) ESRC funded.
  • Large scale server log analytics - Rui Vieira, PhD student funded by Red Hat/JBoss, supervised by Prof Darren Wilkinson
  • Bayesian inference for large linear models - Keith Newman, PhD student, supervised by Prof Darren Wilkinson
  • Bayesian modelling of covariate effects on survival with application to non-Hodgkin's lymphoma - Juliana Iworikumo Consul, PhD student funded by the Petroleum Technology Development Fund (Nigeria), supervised by Dr Malcolm Farrow
  • Bayesian methods for non-Gaussian time series - Muhammad Safwan Bin Ibrahim, PhD student funded by the Government of Malaysia, supervised by Dr Malcolm Farrow
  • Non-Hodgkin's lymphoma prognostic index (CI: Dr Malcolm Farrow) funded by the Scotland and Newcastle Lymphoma Group.

Recent grants, fellowships and studentships

In addition to those listed in the Statistical Bioinformatics and Stochastic Systems Biology group:

  • Mathematical models for the developed Neolithic (PI: Dr Graeme Sarson; CI: Prof Anvar Shukurov, Prof Richard Boys and Dr Andrew Golightly), funded by Leverhulme Trust (2009–12).
  • Imprecision and robustness in experimental design — Noorazrin Abdul Rajak, PhD student funded by Universiti Pendidikan Sultan Idris, Malaysia (2008–2012), supervised by Dr Malcolm Farrow.
  • Predictive dynamic modelling for next generation processing — Javier Serradilla, EPSRC CASE funded PhD student sponsored by BP Oil International Ltd (2008–11), supervised by Dr Jian Q Shi and Prof Julian Morris.
  • Belief representation for counts in Bayesian inference and experimental design — Kevin Wilson, EPSRC funded PhD student (2007–11), supervised by Dr Malcolm Farrow.
  • Bayesian spatio-temporal modelling of rainfall through non-homogeneous hidden Markov models — Sarah Germain, supervised by Prof Richard Boys and Dr Malcolm Farrow, funded by EPSRC. PhD awarded 2010.
  • Bayesian survival analysis for prognostic index development with many covariates and missing data — Xiaohui Zhao, supervised by Dr Malcolm Farrow. PhD awarded 2010.
  • Parallel Bayesian Computation funded by Royal Society (PI: Prof Darren Wilkinson)

Publications

For information on publications, please see the personal webpage of each member of the group.