Dr Sperrin is Senior Lecturer in Health Data Science in the Health eResearch Centre (www.herc.ac.uk), University of Manchester, with a background in statistics.

My primary research interests are in the development of new statistical methodology for prediction, in healthcare applications. Specifically:

1. Role of causality in prediction. Wile causality is typically seen as a separate type of analysis to prediction, it's increasingly recognised that incoporating causality in prediction models brings a number of advantages. For example, the ability to ask 'what if' prediction questions (counterfactual prediction), improving model generalisability and face validity, and the study of fairness. 

2. Adaptive sampling and informative observation/missingness in prediction. Informative observation refers to the fact that presence or absence of data is almost always informative (e.g., in electronic health records). This is both a problem and opportunity, and I am interested in the implications. Adaptive sampling is a proactive question about when to measure things, trading off improved prediction with costs.

3. Model updating and validation. Prediction model performance needs to be checked, and maintained over time and space. 

If you are interested in the above areas, we are always looking for new people to join our group, at all levels. Please get in touch to find out more, making your approach as specific as possible.

Dr Sperrin contributes to the MSc Health Data Science. He also runs training courses for NHS staff and supports undergraduate and postgraduate teaching across the Faculty of Health and Medicine.

Royal Statistical Society, CStat.

FHEA

Statistics, Epidemiology, Data Science

PhD, Statistics, Lancaster University (2010)

MMORSE (Masters in Maths, Operational Research, Statistics and Economics), University of Warwick (2006)