New top story on Hacker News: You need 16 times the sample size to estimate an interaction than a main effect

You need 16 times the sample size to estimate an interaction than a main effect
19 by luu | 0 comments on Hacker News.


Popular posts from this blog