Bayesian analysis of Slate Star Codex survey data.

[Epistemic status: I’m teaching myself Bayesian analysis out of an O’Reilly-esque programming book; I haven’t yet mustered myself to crack the intimidating Andrew Gelman tome on my shelf. I beg you, correct me if I have screwed this up.]

Scott Alexander posted his survey data results several months ago, and recently has been posting some interesting things about how different groups perceive optical illusions.

As part of my quest to finally understand the differences between Bayesian analysis and frequentist analysis, I downloaded his data and poked at it with PyMC, again modeling my analyses after those in chapter 2 of Bayesian Methods for Hackers, by Cameron Davidson-Pilon (the A/B testing example and the Challenger example.)

Continue reading

The same thing, but with different priors

A couple of days ago I posted a Bayesian re-analysis of the data from a paper on prenatal progesterone exposure and sexual orientation.  For that analysis, I used uniform priors for both exposed and unexposed subjects – that is, I assumed we pretty much don’t know anything about how common non-heterosexuality is, and that the effects of progesterone exposure could be anywhere from infinity to nothing.  These priors didn’t seem very realistic, but the results I got seem fairly intuitive, given the data and outside figures on how common non-heterosexuality is.

Continue reading