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

Then you should take the Bayesian side, because Bayesians look at the data first, and they take their data as given rather than taking a null hypothesis as given. They don't just blindly go off and run a test (which assumes a particular prior implicitly that may be wildly inappropriate) and see what it says about the likelihood of their already observed data being generated by the test's assumed data generator.


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: