3 ms·

So you are saying we should abandon evidence-based science when reasoning under uncertainty with asymmetrical outcomes?

• AbrahamParangi 6y ago
  There are many decisions where evidence is near worthless and you have to follow alternative lines of logic. Any circumstance where you’re dealing with low probability, high impact events you will necessarily lack evidence to estimate the true expectation from prior experience (aka evidence). Your prior will be biased in its mean and almost no amount of “evidence” can fix that.
  • riskneutral 6y ago
    Fortunately epidemics are an area where data can be collected until there is no lack of evidence.
    • AbrahamParangi 6y ago
      Epidemics are a great example where we can say that prudence is recommended. Perhaps 80% of the population of the Americas was destroyed by the eurasian plagues - diseases which Europe had reached viral equilibrium with but the Americas had not. The median of a new outbreak is not worrying, the mean is worrying, and the min is apocalyptic.