Statistical Consequences of Fat Tails

This book, the first volume of the Technical Incerto series, investigates the misapplication of traditional statistical methods to fat-tailed distributions.

It highlights the failures of these methods in real-world scenarios, particularly in finance, economics, and behavioral economics.

The book discusses how switching from thin-tailed to fat-tailed distributions requires more than just a change in assumptions, and it explores the consequences of fat tails on statistical inference, including issues with sample means, empirical distributions, parameter uncertainty, and dimension reduction.

It also addresses how many biases in psychology become rational under more sophisticated probability distributions and how most failures in financial economics can be attributed to using the wrong distributions.