He's directly quoting from the paper!
To be fair, the definition of CI is one of those things I try to get into my head every couple of years, and it falls right back out! Also true of some people much smarter than me. (cc: @Meaningness)
-
This Tweet is unavailable.
-
-
Replying to @St_Rev
I read something only about two days ago that said this definition was a common misunderstanding and totally wrong. The only thing I know about statistics is that whatever you think it says, it doesn’t. It always says something that sounds similar but is completely different.
0 replies 0 retweets 4 likes -
This Tweet is unavailable.
-
Yes, that other one is even more wrong. But, consulting wiki, I think the quoted one is also (less) wrong. You have to read the first para really carefully to see the difference. Disclaimer: I don’t know any stats. https://en.m.wikipedia.org/wiki/Confidence_interval …
1 reply 0 retweets 0 likes -
Replying to @Meaningness @St_Rev
If my understanding is correct, the CI doesn’t tell you *anything* directly about data. It’s a property of the *model*, specifically how likely it is that a real-valued parameter of the model is within bounds.
1 reply 0 retweets 0 likes -
Replying to @Meaningness @St_Rev
Everyone reasonably assumes that statistical methods must tell you something you’d want to know (or else why use them?). But they all end up telling you something totally bizarre and useless. So you immediately forget that, and misremember that they do what you’d expect.
0 replies 0 retweets 2 likes -
This Tweet is unavailable.
-
I had to google that. LOL!
0 replies 0 retweets 0 likes -
This Tweet is unavailable.
Thank you! Tonight maybe!
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.