Why is it difficult to predict the evolution of the pandemic reliably?
The present situation is very dynamic. From one day to the next, new regulations are introduced in some regions, different from others’. And in order to have a long-term model, you need the basic stable conditions. Otherwise, whatever model you use, you will not be able to predict the pandemic evolution in the long term. Marina Pollán, director of the National Centre for Epidemiology, recently said that the virus itself is changing. If, in addition to this, contextual conditions change and the data are not easy to measure, reliable predictions cannot be made.
What we can control is the data. We must try not to abandon what we can control and make a collective effort to find out which variables can help us. Perhaps we need to take a look and think about how we can model some variables that do not exactly correspond the data we receive, but rather an approximation, such as excess mortality, but which can help us to better understand what is happening.
Answered by Anabel Forte. Professor of Statistics and Operations Research at the University of Valencia.
«Mètode’s whys and wherefores» is a popular science section in which readers can submit their scientific doubts or questions and an expert on the subject will answer them. You can submit your questions with this form. Among all those we publish, a set of Mètode publications will be drawn every quarter.