News for hackers. Stuff that matters.

A comment on Optimizely Statistics Engine. Parent comment

Re:Optimizely Statistics Engine by darkxanthos on Tuesday January 20, 2015 @10:34AM (#8917346)
It's actually sequential testing which is not necessarily Bayesian. Also, it's not a cheap knock off. Just another way of getting to an answer.
[ Parent | on HN ]
Re:Optimizely Statistics Engine by mziel on Tuesday January 20, 2015 @10:41AM (#8917393)
Bayesian updating of the posterior? Or if you prefer frequentist algorithms for online learning of classifiers?

Not trying to pick a fight, just as a statistitian/ML developer I've seen the same things be reinvented and renamed so many times.

[ Parent | on HN ]
Re:Optimizely Statistics Engine by jmalicki on Tuesday January 20, 2015 @11:05AM (#8917548)
No, if you read their technical paper, it's frequentist sequential testing with false discovery rate control, which is a fairly recent development (I mean, 25 years old is pretty new in statistics).

http://pages.optimizely.com/rs/optimizely/images/stats_engin...

[ Parent | on HN ]
Re:Optimizely Statistics Engine by dlss on Tuesday January 20, 2015 @11:55AM (#8917814)
I think all OP is trying to point out is that it either agrees with bayesian methods or it's wrong... so at best it's not materially new, and at worst it's using questionable assumptions.
[ Parent | on HN ]
Re:Optimizely Statistics Engine by GFK_of_xmaspast on Tuesday January 20, 2015 @11:18PM (#8921420)
"it either agrees with bayesian methods or it's wrong"

This kind of faith-based statistics is pure ideology.

[ Parent | on HN ]