|
|
Some key current areas of focus for statistical theory, methods and computational
research include:
- Statistical modelling with large data sets and high-dimensional covariate
spaces in regression and classification models for prediction;
- Latent structure in scientific modelling, including latent factor models,
including applications in high-dimensional statistical models, multi-scale time series, and others;
- Graphical models, and related statistical and graph theory for graph structuring;
- Multivariate analysis and high-dimensional random graphs and matrices;
- Graphical and factor models in multivariate financial time series and
portfolio analysis;
- Statistical models in "systems biology" - specifically,
modelling and analysis of non-linear, multivariate processes
representing the stochastic dynamics of cellular networks at the single cell level;
- Bayesian prediction tree models and related approaches to flexible non-linear/semi-parametric
regression and prediction;
- MCMC and other simulation methods for Bayesian computation;
- Large-scale model search and stochastic methods for evolutionary exploration
of complex model spaces;
- Statistical computation via cluster computers.
Current collaborations include a major focus on genomics and emerging areas
of systems biology, including:
- Molecular profiling in cancer (and other areas),
prognostic models for predicting clinical
outcomes using molecular profiles, clinical evaluation and testing;
- Identification of complex, multi-dimensional genomic patterns related to
deregulation of oncogenic pathways, and statistical characterisation of
biological pathways using factor models and related methods;
- Parameter estimation and model evaluation using
discrete-time dynamic stochastic models of dynamic cellular networks
- Imaging methodology for single-cell fluorescent imaging
- Statistical mixture modelling of large-scale spatio-temporal organisation
of multiple cell types in immunology
|
