At least one (and probably more like 2-3?) reasonable introductions to Bayesian statistics in practice. I got it from Nate Silver’s The Signal and the Noise + a Coursera course. I also hear good things about Bayesian Data Analysis by Gelman et al, various writings by Yudkowsky, and various lecture videos online. Students currently in university should also consider taking an introductory class in statistics (though supplementary work needed if it’s not Bayesian, which most introductory courses are not).
Some practice is probably also useful.
I suspect the exact material doesn’t matter that much here since the core lesson isn’t that hard, just that the level of statistical rigor that most EAs casually absorb in the “EA memesphere”, while higher than many other places, is nonetheless lower than can be gleaned from an introductory course in Bayesian statistics, and the latter is (I claim) better for the overall culture.
a really great book for learning practical bayesian statistics is richard mcelreath’s statistical rethinking. there is also a series of lectures on youtube.
Agree with this but would note that “The Signal and the Noise” should probably be your first intro or likely isn’t worth bothering with. It’s a reasonable intro but I got ~nothing out of it when I read it (while already familiar with Bayesian stats).
I think it was this course. I don’t have strong opinions on whether it’s better or worse than other introductory courses.
Richard Mcelreath’s lecture videos online feel more fun, but I’m also not sure if this tracks better pedagogy, especially for people exposed to Bayesian stats for the first time rather than using it as a refresher.
Here’s my guess:
At least one (and probably more like 2-3?) reasonable introductions to Bayesian statistics in practice. I got it from Nate Silver’s The Signal and the Noise + a Coursera course. I also hear good things about Bayesian Data Analysis by Gelman et al, various writings by Yudkowsky, and various lecture videos online. Students currently in university should also consider taking an introductory class in statistics (though supplementary work needed if it’s not Bayesian, which most introductory courses are not).
Some practice is probably also useful.
I suspect the exact material doesn’t matter that much here since the core lesson isn’t that hard, just that the level of statistical rigor that most EAs casually absorb in the “EA memesphere”, while higher than many other places, is nonetheless lower than can be gleaned from an introductory course in Bayesian statistics, and the latter is (I claim) better for the overall culture.
a really great book for learning practical bayesian statistics is richard mcelreath’s statistical rethinking. there is also a series of lectures on youtube.
My roommates and I are watching his lectures right now!
Agree with this but would note that “The Signal and the Noise” should probably be your first intro or likely isn’t worth bothering with. It’s a reasonable intro but I got ~nothing out of it when I read it (while already familiar with Bayesian stats).
What’s the coursera course you coursed and do you recommend it?
I think it was this course. I don’t have strong opinions on whether it’s better or worse than other introductory courses.
Richard Mcelreath’s lecture videos online feel more fun, but I’m also not sure if this tracks better pedagogy, especially for people exposed to Bayesian stats for the first time rather than using it as a refresher.