Predictive Monitoring of COVID-19
Updated on May 11
COVID-19 predictions are foundamentally important for rationalizing planning and mentality, but also challenging due to the innate uncertainty of the complex, dynamic and global COVID-19 pandemic as a typical wicked problem. Traditional prediction or forecasting efforts, which aim to make an accurate prediction now to come true in the future, might be misleading in this context of extreme uncertainty and dynamic changes. To deal with uncertainty we proposed predictive monitoring of the epidemic life cycle curves together with accumulating actual data, to capture changes in connective predictions, which would be traditionally viewed as bad or proof of failure of a prediction model, and make sense such changes as meaningful siganls of uncertainty and changes in the real-world scenarios. In turn, such signals from the predicted theoretical future may inform, initiate, and guide actions now to influence the real future. Motivation, theory, method, cases, and caution are in this white paper.
Disclaimer: Content from this website is STRICTLY ONLY for educational and research purposes and may contain errors.