Never produce a point estimate for risk management, esp. in a fat tailed domain, rather show statistical properties.
Never judge a risk management stance from point forecasts.

These errors are promoted by the Idiot of Pennsylvania, Phil the rat 🐁 Tetlock @PTetlock.
Paper with @DrCirillo forthcoming in Nature Physics later this week explaining the point, @mattwridley. Tail exponent <1, even possibly <½ means NEVER issue a single point forecast and NEVER present an average.
For pandemics if the average exists, about ~95-99.9% of observations would fall BELOW the average. Aside from the slow workings of the law of large numbers.

Should I explain this in a short video?
You can follow @nntaleb.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: