N N Taleb's Probability Questions (UNOFFICIAL) Universa's Bernoulli for Portfolio Simulation: Correcting the Empirical Distribution (2024)
Nov 11, 2025
Brandon, a representative from Universa, and Ron, an institutional portfolio expert, dive into the innovative Bernoulli portfolio-simulation tool. They explore how Bernoulli maximizes geometric returns and the flaws of traditional empirical distributions. Key discussions include tail extension strategies for unseen extreme events and the surprising benefits of zero-return puts. Ron showcases Bernoulli's stress-testing features, emphasizing its application in complex institutional portfolios, aiming to redefine risk management and compounding efficacy.
AI Snips
Chapters
Transcript
Episode notes
Prioritize Compound Growth
- Maximizing geometric return (compound growth) should be the investor's primary objective.
- Optimizing log-returns naturally produces strategies that avoid large losses and mimic risk aversion.
Stress Test Surprise At First Boston
- Nassim recounts a First Boston stress-test that used the worst historical drawdown and got surprised by a larger crash.
- A record is always exceeded, so using past maxima/minima underestimates true risk.
Correct The Empirical Tail
- The empirical distribution underestimates tail risk because samples miss rare extreme events.
- Use extreme value theory to extend tails and estimate realistic minima/maxima over windows.