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

> ChatGPT punched out a reasonably sensible t test!

I think the distribution is decidedly non normal here and the difference in the medians may well have also been of substantial interest -- I'd go for a Wilcox test here to first order... Or even some type of quantile regression. Honestly the famous Jonckheere–Terpstra test for ordered medians would be _perfect_ for this bit of pseudoanalysis -- have the hypothesis that M3 > M2 > M1 and you're good to go, right?!

(Disclaimers apply!)



12,000 builds? Sure maybe the build time distribution is non-normal, but the sample statistic probably is approximately normal with that many builds.


Many people misinterpret what is required for a t-test.


I meant that the median is likely arguably the more relevant statistic, that is all -- I realise that the central limit theorem exists!


Fair enough - I see what you are saying. Sorry for my accidentaly condescending reply.




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

Search: