Forecasting for COVID19 has Failed
DSA ADS Course - 2021
Forecasting, COVID19, John Ioannidis, Public Policy, Health Policy, Causal Inference, Forensic Medicine, Causality, Intuitive Causation, Probabilistic Causation
Forecasting is usually impossible in high causal density environments. Scenario planning with applied probability and adaptation to near real-time data is optimal strategy. Epidemic forecasting is usually a fools errand yet appropriate analysis of experience and historical precedent is helpful.
In scenario planning consider attempt to find true causality - intuitive causation vs. probabilistic causation. Probabilistic causation is optimal yet discuss other techniques to seek true causation. Sometimes true causation is impossible to find and discuss optimal strategy under such circumstances.
Epidemic forecasting has a dubious track-record, and its failures became more prominent with COVID-19. Poor data input, wrong modeling assumptions, high sensitivity of estimates, lack of incorporation of epidemiological features, poor past evidence on effects of available interventions, lack of transparency, errors, lack of determinacy, looking at only one or a few dimensions of the problem at hand, lack of expertise in crucial disciplines, groupthink and bandwagon effects and selective reporting are some of the causes of these failures. Nevertheless, epidemic forecasting is unlikely to be abandoned. Some (but not all) of these problems can be fixed. Careful modeling of predictive distributions rather than focusing on point estimates, considering multiple dimensions of impact, and continuously reappraising models based on their validated performance may help. If extreme values are considered, extremes should be considered for the consequences of multiple dimensions of impact so as to continuously calibrate predictive insights and decision-making.