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

But they already do this. When they design clinical trials, they take type 1 and type 2 errors into account. Even if you wanted to accept a ton more error, which would mean a lot more bad drugs on the market and a lot more good drugs off the market, it's still not going to be cutting costs by orders of magnitude or something.


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

Search: