Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Yeah. Calling something a five-sigma event loses meaning if there haven't been anywhere near enough time intervals (here: market days) to validate that thesis.

Here is a decent paper on the topic of n-sigma events: https://arxiv.org/pdf/1103.5672.pdf



Not just that, you have to define the independent units. A 1 in 35M event is crazy over 35M days. It's less crazy over 35M trades. Without knowing which it is, 1 in 35M doesn't say anything.




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: