Hacker News new | past | comments | ask | show | jobs | submit login

“a significance level of > 20σ”

Insane! Clearly that doesn’t include the uncertainty in our understanding of physics or neutron stars.

Edit: I tried to work out the % that 20σ is, but it is so mind bogglingly small that there should be a law against using such an insane number in any serious context.




Each tail is bounded above by e^(x^2/2) / (x sqrt(2 pi)). Take logs and you get (x^2 / 2) - log(x) - log(2 pi)/2. Calculate that out and you get to about 1.44 * 10^85. You can double that for 2 tails. But it is still an impressively small number.




Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: