Nassim Nicholas Taleb: The Black Swan: The Impact of the Highly Improbable
ANNOUNCEMENT: We are looking for a civilized public debate with two or three academic detractors, preferably persons with “strong credentials” in their milieu (say, an analytical philosopher with known and well cited works who disagrees with my “inseparability” between chaotic pseudorandom and purely random, a “prominent” economist deemed to have what academics call “a publication record” who does not share my opinion about his profession, etc.). Good debating manners required –See the smooth C SPAN debate with Charles Murray. Ideally the debate would be televised. Please write to gamma [at] fooledbyrandomness [dot] com.
Critique by the American Statistical Association (forthcoming) and my reply in The American Statistician (in press, August 2007). Addendum: My Poisson Buster: Why Poisson does not work out of sample (14 Million pieces of data showing Poisson failures and unavailability of a characteristic scale).
T-Shirt—a gift from Peter Bevelin: Statisticians have a hard time figuring out the difference between Absence of Evidence and Evidence of Absence outside of exam questions.
Comment about Tyler Cowen’s discussion of derivatives as predictors and technical appendix -- Unlike other, more technical critics, I do not think much of Cowen’s intellect, abilities, & understanding of probability & random payoffs, but that irresponsible fool was the first to advertise the contribution of “prediction markets” in high moment applications, heavy-tailed environment. “Prediction markets” fail in fat-tailed domains because of estimation errors. Also note a blogger who got my point about predicting in Extremistan. Update: Note that Cowen had the talent to argue that markets know how to adequately price remote events right ahead of the subprime debacle.
Note on a common error: from my Lectures in the philosophy of probability.