I work in the area of computational statistics. I am currently working collaboratively on spatio-temporal statistical methods, associated computational algorithms and the integration of these into web-based information systems.
Methodological and computational aspects of log-Gaussian Cox Processes: I have authored an R package for Bayesian analysis of various kinds of log-Gaussian Cox processes.
Combining data recorded at multiple spatial scales: Often epidemiological exposure and outcome data is collected on different spatial units. I am interested in combining this information in a principled manner
Filtering methodology and applications:Particle filters are an advanced methodology for inferece with sequentially collected data, during my PhD I worked on an adaptive Sequential Monte Carlo Smapler. I am keen to use these methods where I can in my research. An example of where I have applied these methods is in the forecasting of meningitis incidence, below.
Forecasting meningitis incidence in sub-Saharan Africa in collaboration with the World Health Organisation: In this project, we provide WHO with real-time forecasts of Meningitis incidence in sub-Saharan Africa.
Spatial modelling of survival outcomes: This is joint work with my PhD student Emmanuel Singogo. We will be investigating new methods for analysing spatially-referenced survival data. We will be applying these methods to help understand survival patterns in a group of co-infected HIV/cancer patients in the Zomba region of Malawi.
