超过 2 次分割的分割检验的显着性检验
对于两次以上实验的百分比指标,您应该使用什么显着性检验?
例如,
Version | Clicks | Impressions
A | 5 | 1,763
B | 4 | 1,672
C | 2 | 1,689
我们如何确定版本 A 确实优于其他两个版本?
What significance test should you use for a percentage metric with more than two experiments?
For example,
Version | Clicks | Impressions
A | 5 | 1,763
B | 4 | 1,672
C | 2 | 1,689
How sure are we that verison A really is superior to the other two?
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(1)
过去,我个人在顶部和底部之间进行了成对的 G 测试,将置信度乘以 n select 2 的模糊因子,以考虑到存在 n select 2 可能的对可能是最极端的事实。从理论上讲,这过于保守,但它对我有用。
请参阅http://elem.com/~btilly/ effective-ab-testing/ 了解更多。
In the past I personally have done a pairwise G-tests between the top and the bottom, multiplying the confidence by a fudge factor of n choose 2 to account for the fact that there are n choose 2 possible pairs that could have been the most extreme. Theoretically this is overly conservative, but it worked for me.
See http://elem.com/~btilly/effective-ab-testing/ for more.