The Exeter Initiative for Statistics and its Applications (ExIStA) held a series of member presentations at the University Exeter on Thursday 5th May.
The event comprised talks from four speakers on a variety of different themes, with Lucy Knight from Devon County Council, co-founder of the Open Data Institute Devon Node, kicking off the proceedings. Lucy talked about the challenges faced with realising the potential benefits of open data and how Devon County Council are championing the drive to open up access to datasets from the county and generate meaningful insights from these data.
Next we heard from Laura Dawkins who recently completed her PhD viva in the Exeter Climate Systems group at the University of Exeter – congratulations Laura! Her research included an investigation of decreases in losses (i.e., insurance claims) due to European windstorms. She explained how storms have been weaker and therefore less damaging in recent years and that this is likely due to a change in the North Atlantic Oscillation – a weather phenomenon characterised by a north-south shift in the track of storms across the North Atlantic Ocean and into Europe. This work has recently been submitted for publication in the journal Natural Hazards and Earth System Sciences: The 21st Century Decline in Damaging European Windstorms.
Tim Paulden, Innovation & Development Manager at ATASS Sports, gave us an insight into how statistical models can potentially be used to detect match-fixing in tennis. By utilising in-play data on the betting market odds and comparing this with the “expected” match win probability under normal circumstances throughout the match, Tim explained how discrepancies might be used to highlight suspicious market movements and therefore potential betting irregularities. ATASS have recently funded a PhD student to look into this area in more detail over the next three years.
Last but not least, Kate Martin, a medical statistician from NHS Blood and Transplant, presented some fascinating research into the effect of out-of-hours transplantation on liver transplant outcomes. Using data from across the seven liver transplant centres in the UK, over fourteen years, survival was compared for weekday versus weekend transplants, and for daytime versus night-time transplants. Accounting for various clinically-advised, potential confounders, somewhat surprisingly there was evidence of a positive weekend effect, i.e., weekend transplants were found to be beneficial to graft and transplant survival at 30 days for liver recipients. No evidence was found for a difference in survival between daytime versus night-time transplants. Potential explanations for the weekend effect were discussed including, the possibility of better post-transplant care on a well-staffed ward early in the week; centres being risk-adverse at weekends; and there being fewer routine elective commitments and lower levels of routine and complex work undertaken within theatres and intensive care units for staff to deal with on weekends.
The diversity of the talks once again highlighted the variety of areas in which statistics are being used within local public and private sector organisations. We rounded off the event with tea and cake and it was great to see so many people taking the opportunity to network with other ExIStA members.
Look out for details of future events on the ExIStA website.