The Mudcat Café TM
Thread #167636   Message #4046785
Posted By: SPB-Cooperator
18-Apr-20 - 05:52 AM
Thread Name: BS: Coronavirus statistics
Subject: RE: BS: Coronavirus statistics
The only 'safe' predictive models are those that project trends based upon empirical data. The biggest problem is that the modelling has to take into account a feedback loop that takes into account of how mass behaviours change reactively as the trends progress. Over a long enough time frame that also can be predictable. The biggest curveball is the impact of proactive measures, these can be predicted by overlaying trends of localities that has different measures in place.