[ / / / / / / / / / / / / / ] [ dir / random / 93 / biohzrd / hkacade / hkpnd / tct / utd / uy / yebalnia ]

/ratanon/ - Rationalists Anonymous

Remember when /ratanon/ was good?

Name
Email
Subject
REC
STOP
Comment *
File
Password (Randomized for file and post deletion; you may also set your own.)
Archive
* = required field[▶Show post options & limits]
Confused? See the FAQ.
Embed
(replaces files and can be used instead)
Oekaki
Show oekaki applet
(replaces files and can be used instead)
Options

Allowed file types:jpg, jpeg, gif, png, webp,webm, mp4, mov, pdf
Max filesize is16 MB.
Max image dimensions are15000 x15000.
You may upload5 per post.


File: 9fd5da22f16f7f0⋯.jpg (420.2 KB,600x972,50:81,Venn-J-IM-Maull-004717[1].jpg)

 No.9656

someone redpill me on frequentist probability pls

____________________________
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9657

Frenquentism is bayesianism looked differently, it's fine.

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9658

>>9657

HERESY!

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9659

how else would you predict the probability of an unprecedented event

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9660

>>9659

P R I O R S

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9661

>>9659

Principle of indifference on the space of all possible events.

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9662

>>9661

The naive principle of insufficient reason depends on your arbitrary individuation of states. Other forms of uninformative prior distributions have stronger theoretical grounding, e.g. Jeffreys' prior.

Besides, that is all still about Bayesian prior distributions and not really "frequentist" in any way at all. This is supposed to be a thread about frequentism.

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9663

>>4035

Do you even know anything about statistics besides Yud's BS in the sequences?

>Solomonoff induction is literally uncomputable and therefore worthless

>using the $CURRENT_YEAR as an argument

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9664

>>9663

>Yud's BS

He doesn't have one, he dropped out of high school

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9665

Frequentism deals with probabilities that have an objectively correct value in some sense–given the assumption we live in a universe where everything happens according to mathematical laws, there must be some objective truth about what statistics you would get if you could repeat any given experiment a near-infinite number of times from the same (or randomly perturbed) starting conditions. That's why it's better to assume frequentism when interpreting probabilities that appear in foundational theories like quantum physics (where probabilities are the frequencies you'd get if you repeated the same quantum experiment a near-infinite number of times), statistical mechanics (where probabilities are the frequencies you'd get for different outcomes if you prepared a near-infinite number of systems with identical macro-variables like the total energy of a system or its volume, but randomly varying microscopic conditions) or neo-Darwinism (where the fitness associated with a given value of a trait is the fraction of organisms who'd survive and reproduce if you created a near-infinite number of organisms with that value of the trait and put them all in identical environments; this incidentally is why creationists are wrong to say the definition of fitness is circular, since the actual number of organisms with that trait may not be near-infinite so the frequency that actually survive and reproduce may be different from the fitness, just like how if you have a fair coin, which would have a 50-50 ratio of heads to tails if you flipped it a near-infinite number of times, but you only flip it say 10 times, the ratio may be different than 50-50).

On the other hand, Bayesianism (separate from Bayes' equation, which is used in both Bayesian and frequentist approaches) is about subjective rather than objective probabilities, for example a Bayesian could assign a "probability" to an unproven mathematical statement like P!=NP that they might update as they learned new things, whereas that wouldn't make any sense in frequentist terms. I also get the sense that Bayesianism is usually more useful in a practical sense when you aren't just trying to come up with theoretical models but are trying to do hypothesis testing with empirical evidence, see http://www.fharrell.com/2017/02/a-litany-of-problems-with-p-values.html for some arguments as to why it's better than the usual frequentist approach of trying to find p-values of a given observation assuming some null hypothesis.

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9666

They're just two different incompatible definitions of "probability". No contradiction, just two separate things with the same name but minimal overlap. Like "African-Americans" and real people.

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9667

>>9666

Or goyim and real people.

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.

 No.9668

Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.



[Return][Go to top][Catalog][Nerve Center][Random][Post a Reply]
Delete Post [ ]
[]
[ / / / / / / / / / / / / / ] [ dir / random / 93 / biohzrd / hkacade / hkpnd / tct / utd / uy / yebalnia ]