Lecture | online
Corona statistics, Corona uncertainty
- Thursday 7 May 2020
- Virtual lecture room NGL
In deciding their policy in this Corona crisis, the Dutch government is guided by what the experts say. But how reliable are the conclusions these experts draw with the state-of-the-art mathematical models at hand? In an online lecture for Natuurwetenschappelijk Gezelschap Leiden Professor emeritus of Mathematical statistics Richard Gill will explain how RIVM's predictions come about. And which model inadequacies we have to take into account.
The inadequacies of models
The RIVM uses a refined version of the standard four-state compartment model for epidemics ('SEIR') to make predictions about the result of policy interventions like social distancing or quarantining. Just like in any model, there are various known inadequacies. It is therefore important to appreciate these inadequacies and understand and quantify the uncertainties involved in the calculations and reached conclusions.
As a statistical mathematician, Professor Richard Gill closely follows the work of the RIVM scientists and his conclusion is that they are rapidly learning a great deal more about the inadequacies of present 'state-of-the-art' modelling. Gill: 'More than ever we are aware how hard it is to communicate uncertainty to the public and to decision-makers. Someone with a hammer sees nails everywhere which need to be hammered. This is painfully obvious when one looks at applied mathematicians, statisticians, computer scientists, physicists … all with their own little hammers.'
About Richard Gill
In 1974, after studying Mathematics in Cambridge, UK-born Brit Richard Gill moved to the Netherlands to pursue a PhD degree at Mathematical Centre (MC) in Amsterdam. After various posts at the Dutch Free University, MC and Leiden University, he became a full professor of mathematical statistics in Utrecht. In 2006 he transferred to Leiden University. He reached the mandatory retirement age in 2017, and continues with research and consultancy. Gill is most well-known for his research on counting processes and survival analysis.
Natuurwetenschappelijk Gezelschap Leiden in times of Corona
In a series of short, high-end lectures Leiden scientists will discuss various aspects of the Corona crisis from their scienctific expertise. So join us every two weeks in our virtual NGL lecture room at 19.45! A few of these lectures will be held in English.