Statistical Consequences of Fat Tails
TL;DR — The technical volume behind the Incerto — what actually breaks when the variance is infinite, with the maths left in.
⬇ Download the archived copy — kept here so the document survives its source going dark. arXiv preprint (author distribution licence).
523 pages, free from the author. The Technical Incerto volume: the formal treatment of everything the popular books assert. Where The Black Swan argues that standard statistics misleads under fat tails, this shows exactly which estimators fail, how fast, and what to use instead.
The practical core is that the sample mean stops being informative long before most practitioners suspect — for many real distributions, the number of observations needed for the law of large numbers to bite is orders of magnitude beyond any dataset in hand. That is a claim about portfolio construction, not just statistics.
The most substantial single document recovered from either thread, and the one with the most direct bearing on Capital Allocation.
Where this came from
523 pages. A copy is archived locally against link rot; the header links the original source.