Tag: Fat Tails

[Twitter | X] Nassim Nicholas Taleb Conjecture Tested: Experiment Shows Results of Trading with News in Advance

On By In Academic, Economics, Twitter, X

Link to paper – papers.ssrn.com/sol3/papers.cfm?abstract_id=4965616

[YouTube] Universa’s Bernoulli for Portfolio Simulation: Correcting the Empirical Distribution

On By In Economics, Videos

The empirical distribution is not empirical, full of Turkey problems. “Real tails” do not show in past samples because of their property under fat tails. Empirical distributions are, by design, interpolating; we fix by extrapolating and extending the tails. Note the dispute: In French, Bernoulli is pronounced “Ber-noo-yi” not “Ber-noo-li”, much like consigliere is pronounced “consiyyere”.

[YouTube] Why Correlation is Unreliable

On By In Statements, Videos

At the 2022 Greenwich Economic Forum-Miami, Black Swan author, Nassim Nicholas Taleb explains why correlation is unreliable as a due diligence tool. Coming as it does during an ongoing pandemic and in the middle of Vladimir Putin’s invasion of Ukraine, Taleb also discusses Wars and Pandemics and puts them into their proper risk buckets.

[YouTube] Claims that Violence Has Dropped Are Not Statistical

On By In Statements, Videos

Violence is from Extremistan, hence requires some more sophisticated tools since LLN works slowly. We see how Pinker’s thesis is bogus. We look at ways to integrate the factual unreliability of historical accounts. We look at transformations to analyze violence using Power law tools since the worst case is bounded at the contemporary population level.

Links to Papers with Pasquale Cirillo

On the statistical properties and tail risk of violent conflicts [TECHNICAL, PHYSICA A]

The Decline of Violent Conflicts: What Do the Data Really Say? [NONTECHNICAL, NOBEL FOUNDATION]

[YouTube] P-Value Hacking

On By In Videos, Writing

We saw that 1) many metrics are stochastic, 2) what is stochastic can be hacked. This is the simplification of my work showing that “p-values are not p-values”, i.e. highly sample dependent, with a skewed distribution. For instance, for a “true” P value of .11, 53% of observations will show less than .05. This allows for hacking: in a few trials, a researcher can get a fake p-value of .01.

Paper is here and in Chapter 19 of SCOFT (Statistical Conseq of Fat Tails): Link to paper – A Short Note on P-Value Hacking