Professor Theo Economou

I'm a statistical data modeller, developing and appying models in the area of environmental epidemiology. I have a particular interest in climate change and health and a strong background n weather and climate applications. Current topics include:

• Models for the synergy between temperature and air quality on human health
• Integrating probabilistic health risk predictions into early warning systems
• Statistical correction of communicable disease data (disease nowcasting)
• Climate projections of heat related impacts (e.g., school closures in England)
• Data blending of environmental data such as climate output

By modern standards my expertise is in interpretable and explainable supervised learning and AI. My research broadly focuses around Bayesian hierarchical modelling. More specifically methodological research interests include:

• Hierarchical distributed lag models
• Models for correcting disease data
• Space-time modelling using Bayesian hierarchical splines
• Stochastic weather generators
• Downscaling of weather and climate data
• Hidden Markov and semi-Markov models

PhD Opportunities:

• Leveraging probabilistic AI for heat-related health risk mitigation and adaptation (co-supervised by Ben Youngman, Met Office)
• Novel statistical AI approaches for modelling and evaluating extreme windstorm risk (co-supervised with David Stephenson, Adam Scaife, Matthew Priestley, Willis Towers Watson)
• Extreme value modelling of rainfall from high resolution radar data in a changing climate (co-supervised with Ben Youngman, Met Office)
• Robust and scalable methods for hierarchical models in environmental epidemiology (co-supervised by Chunbo Luo)
• Mitigating Climate Impacts towards a national warning platform for reducing heat-related risk on human health (co-supervised by Met Office)
• New statistical and machine-learning methods for forecasting high-impact weather events in urban areas (co-supervised with Stefan Siegert)