Shared by Woolwit
Another paper written by NNT. PDF available for download at link location.
Errors, Robustness, and the Fourth Quadrant
Nassim Nicholas Taleb
NYU-Poly Institute; London Business School
February 14, 2009
Abstract:
The paper presents evidence that econometric techniques based on variance- L2 norm are flawed -and do not replicate. The result is un-computability of role of tail events. The paper proposes a methodology to calibrate decisions to the degree (and computability) of forecast error. It classifies decision payoffs in two types: simple payoffs (true/false or binary) and complex (higher moments); and randomness into type-1 (thin tails) and type-2 (true fat tails) and shows the errors for the estimation of small probability payoffs for type 2 randomness. The Fourth Quadrant is where payoffs are complex with type-2 randomness. We propose solutions to mitigate the effect of the Fourth Quadrant based on the nature of complex systems.