Read on Twitter
ERRORS 101
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. https://twitter.com/mattwridley/status/1259898818017058818
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. https://twitter.com/mattwridley/status/1259898818017058818
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?
Should I explain this in a short video?
Voila
