Predictive Monitoring of COVID-19

4 May, 2020

When will the coronavirus pandemic end?

Data scientists have attempted to answer this question. Their predictions use a mathematical model known as SIR (susceptible - infected - recovered), which calculates the spread and recovery of diseases.

Researchers from Singapore University of Technology and Design (SUTD) fed the model data on confirmed infections, tests conducted, and deaths recorded, to estimate the life cycle of COVID-19.

Globally, their system predicts the pandemic will end this December. But the end date is estimated to vary immensely among nations, from June in Australia to October in Italy.

The site developed provides continuous predictive monitoring of COVID-19 as a complement to traditional monitoring or traditional prediction practices. SIR model is regressed with daily updated data from different countries to estimate the pandemic life cycle curves and theoretical ending dates, with codes from Milan Batista and data from Our World in Data. The continuously updated predictions with the latest data are expected to change as the result of the changes in real-world scenarios over time. Monitoring such changes in the predictions of untestable theoretical future events is aimed to sense uncertainty (indicated by volatility) and detect dynamic changes in present real-world scenarios. Motivation, theory, method, examples, and caution are in this paper. Below are other and more systematic COVID-19 forecasting efforts around the world.