School of Mathematics, Statistics and Operations Research

Statistics

Research in the Statistics Group covers a wide range of areas in both theoretical and applied probability and statistics.

Applied Statistics

Our applied statistics projects include research into methods of estimating the sizes of finite population through capture-recapture modelling, the design and analysis of experiments, multivariate statistics and influence analysis.

In categorical data analysis we are researching the analysis of multiple responses and the development of appropriate diagnostics.

Our research interests in time series analysis involve projects in forecasting that include the development of tests for improved forecasting performance, endogenous detection of structural changes and the development of new evolutionary models for stochastically evolving seasonal patterns.

We are applying Bayesian statistical methods to problems in astronomy and geophysics, providing improved solutions to estimation problems which had previously been intractable or only partly solved. Our research in stochastic point processes is being applied to models of the occurrence of earthquakes.

We have an active research programme in the asymptotic theory of empirical and point processes, martingale methods and differential geometry in statistics. Applications of this research can be found in goodness of fit diagnostics, distribution free methods in regression, the spatial analogue of the change-point problem, insurance mathematics and mathematical and historical demography.

Collaborations

Staff in the Statistics Group have strong links to other research groups in the School, as well as the the following Research Centres: