So I might not have a great enough understanding but isn't 3.8 sigma not high for this kind of announcement (with the current title)? if someone wants to report finding a planet in the solar system ten times the size of Earth wouldn't they do so at more than @3.8 sigma. In particular, the prior is quite low because it would affect a bunch of other astronomical calculations, wouldn't it? So you would want the confidence that much higher....
What I mean is that in human terms, Google says the average man is 5'9" with standard deviation 2.9 inches, then 3.8 sigma (the chances of this happening by chance) are the same as the chances of finding someone 6'8". It doesn't seem unbelievable - https://answers.yahoo.com/question/index?qid=20101208183308A...
If a woman is set up on a blind date and told the guy is 6'8" would she find it like not credible?
My understnading of statistics then is that the idea that there just so happens to be a planet ten times the size of Earth in our solar system is pretty unbelievable. So I'd want the chances of it being an experimental fluke to be lower than seeing a 6'8" guy, anywhere, at any time, in any context. Because I would think that's what happens when you look at enough data.
- Aren't the chances of this being completely spurious literally exactly the same as a researcher saying "I saw a 6'8" guy somewhere", meaning anywhere, in any context?
Wouldn't you want higher evidence before stating "evidence found for ninth planet in our solar system, ten times the mass of Earth"?
No, this is just about right. The evidence they have points sharply to there being a ninth planet. Now they're saying "here's where it might be, we need help finding it!"
Once someone has a photo and an orbit, that'll be the >>3.8sigma you want, but I don't think anyone is saying anything like that yet. What they're saying is "dedicating 5 years of the pinnacle of your planetary astronomy career to looking for this thing is probably not a waste." They directly compare this to the theoretical prediction of Neptune, so I think the authors have this same interpretation of what they are doing.
Before your comment, I edited it to be much more civil and less presumptuous, my original phrasing had included "3.8 sigma is fuck-all", I was briefer, more resolute, and had just included my first link. (You can see evidence of my original phrasing if you remove the hedging I added.)
But the downvotes meant I'm probably wrong - rather than delete I edited it to be much nicer so I could find out why I'm wrong.
Upvoted for tempering the aggression and because I agree with you. As far as I can tell your interpretations of the statistics are correct. I don't see how an argument can include information stronger than the statistics of indirect detection of the object (without direct detection).
I'm having trouble following your height example because "finding" is not well defined, and it's conflating rate of occurrence with certainty.
Being on a blind date with someone who is 6'8" is like knowing someone that won the lottery. The chance of winning the lottery is very low, but lots of people play it and chances are the winner has friends. You're unique, but have no reason to be incredulous.
This data being a coincidence is more on par with one person getting one chance to correctly predict (some of) tomorrow's lottery numbers. It's 1 in 13822.
http://www.dummies.com/how-to/content/how-to-calculate-six-s...
What I mean is that in human terms, Google says the average man is 5'9" with standard deviation 2.9 inches, then 3.8 sigma (the chances of this happening by chance) are the same as the chances of finding someone 6'8". It doesn't seem unbelievable - https://answers.yahoo.com/question/index?qid=20101208183308A...
If a woman is set up on a blind date and told the guy is 6'8" would she find it like not credible?
My understnading of statistics then is that the idea that there just so happens to be a planet ten times the size of Earth in our solar system is pretty unbelievable. So I'd want the chances of it being an experimental fluke to be lower than seeing a 6'8" guy, anywhere, at any time, in any context. Because I would think that's what happens when you look at enough data.
- Aren't the chances of this being completely spurious literally exactly the same as a researcher saying "I saw a 6'8" guy somewhere", meaning anywhere, in any context?
Wouldn't you want higher evidence before stating "evidence found for ninth planet in our solar system, ten times the mass of Earth"?