Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Bayes' Theorem with Lego

countbayesie.com

Post reply on HN
Print view View original

11–19 of 19 posts

Re: Bayes' Theorem with Lego

#11

Post by pdkl95 »

rndn wrote: post #8
ibebrett wrote: post #5

I think it should be pointed out that bayesian's believe bayes theorem as a statement about plausibility of statements, so intuitions about an event space (which are obvious statements about sets) don't carry over to the general case.

Do you (or someone else) happen to know of a good introduction into Bayesian's intuition about the plausibility of statements?

http://jakevdp.github.io/blog/2014/03/11/frequentism-and-bay...

This a very good 4 part series on the subject, with a focus (esp. in part 3) on how this affects scientific statements.

Re: Bayes' Theorem with Lego

#14

Post by ygra »

Dear people who write math: Don't write text as-is in math mode. Either use \mathrm/\mathit (for LaTeX) or "" (for Word) to prevent random text being interpreted (and typeset as) variables adjacent to each other. The spacing for $numberOfYellowPegs$ is particularly horrifying.

Re: Bayes' Theorem with Lego

#16

Post by tmoertel »

rndn wrote: post #8
ibebrett wrote: post #5

I think it should be pointed out that bayesian's believe bayes theorem as a statement about plausibility of statements, so intuitions about an event space (which are obvious statements about sets) don't carry over to the general case.

Do you (or someone else) happen to know of a good introduction into Bayesian's intuition about the plausibility of statements?

The best reference is Jaynes's Probability Theory: The Logic of Science. The first three chapters are online for free and explain the development of probability theory as a logic for plausible reasoning [1]. If you want to get a quick taste, I wrote a short blog post that introduces the theory while working through an interesting coin-toss problem [2].

[1] http://bayes.wustl.edu/etj/prob/book.pdf

[2] http://blog.moertel.com/posts/2010-12-20-more-on-the-evidenc...

Re: Bayes' Theorem with Lego

#17

Post by ibebrett »

tmoertel wrote: post #16
rndn wrote: post #8

Earlier quoted context omitted.

Do you (or someone else) happen to know of a good introduction into Bayesian's intuition about the plausibility of statements?

The best reference is Jaynes's Probability Theory: The Logic of Science . The first three chapters are online for free and explain the development of probability theory as a logic for plausible reasoning [1]. If you want to get a quick taste, I wrote a short blog post that introduces the theory while working through an interesting coin-toss problem [2]. [1] http://bayes.wustl.edu/etj/prob/book.pdf [2] http://blog.moe…

If you hadn't beat me to it, this is exactly what I would have suggested

Re: Bayes' Theorem with Lego

#19

Post by JadeNB »

ygra wrote: post #14

Dear people who write math: Don't write text as-is in math mode. Either use \mathrm/\mathit (for LaTeX) or "" (for Word) to prevent random text being interpreted (and typeset as) variables adjacent to each other. The spacing for $numberOfYellowPegs$ is particularly horrifying.

> use \mathrm/\mathit (for LaTeX)

Even better, use `\text`, which re-sizes properly.

Post reply on HN