New statistical methods that can also be applied beyond the field of mathematics – for example in psychology, medicine or computer science – are the focus of a three-day joint event organised by the Centre for Statistics in Oldenburg and Bremen (ZeSOB). Registrations are open through 2 September.
The ZeSOB “Methods Week” will take place from 9 September through 11 September, this time at the University of Bremen campus. Students, PhD candidates, researchers and anyone else interested in new mathematical methods are invited to attend – particularly those from the Universities of Oldenburg and Bremen. The event is funded by the “NoW – Northwest Alliance Research Fund”.
The event topic describes something that many researchers would surely like to have when exploring scientific questions, as would people in general who have to make decisions over a longer period of time: “Safe anytime-valid inference (SAVI)”.
There have been significant developments in this field at an international level in recent years, particularly in the Netherlands, North America and England, but there are also contributions from Bremen, says Bremen-based statistician Prof. Dr. Thorsten Dickhaus. As evidenced by an increasing number of workshops, publications and the first job vacancies, the SAVI method is becoming increasingly established, adds the Oldenburg-based statistician Prof. Dr. Peter Ruckdeschel. “The aim of this method is to control and keep the error rate low in the long term when dealing with many decisions over an extended period.”
One exampe is the monitoring and treatment of an unconscious patient in hospital based on the monitoring of various parameters such as blood pressure, heart rate and blood values. “If one were satisfied with an error rate of five per cent in this context and were to check the patient’s readings 20 times, the errors could, in the worst-case scenario, accumulate – which, of course, must be avoided,” says Ruckdeschel.
In this area, statistics have long offered “a very limited toolkit”, according to Ruckdeschel; now, SAVI as a method enables consistent, long-term error control. It also helps, for example, banks that wish to avoid liquidity problems at all times.
The speaker at the event is the Dutch mathematician Dr. Alexander Ly, who completed his PhD at the University of Amsterdam in 2018 and is now a researcher at the Centrum Wiskunde & Informatica in Amsterdam. According to host Dickhaus, Ly is characterised by a particular rigour in deriving statistical methods, “combined in a unique way with a very keen eye for efficient implementations and real-world applications, such as those used in manufacturing”. When it comes to implementation and application, Ly is regarded as “one of the world’s leading experts”.
Ruckdeschel emphasises that the three-day ZeSOB event is traditionally designed not only to convey the subject matter but also to offer a “hands-on” opportunity to present and discuss one’s own mathematical or scientific problems. A few spots – including some for online participation, if required – are still available; registration is open via the ZeSOB website through 2 September.
