Category: Announcements and Resources, a.k.a. BayesWatch

Annotated reading list for getting into Bayes

This recently published paper provides an annotated reading list for learning about Bayesian modeling. From one of the authors:

We were invited to submit this paper for a special issue on Bayesian statistics for Psychonomic Bulletin and Review. Each paper in the special issue addresses a specific question we often hear about Bayesian statistics, and ours was the following:

I am a reviewer/editor handling a manuscript that uses Bayesian methods; which articles should I read to get a quick idea of what that means?

Read the rest of the blog post here (also includes a link to the paper).

CASE is offering Bayesian modeling workshops this term!

Just because Bayes Club isn’t meeting doesn’t mean you can’t get a little bit of the old Reverend and his theorem in your life this term!

The first workshop is Oct 23-24, and the second is Nov 6-7. You’ll get tons of focused instruction on how Bayesian modeling works and when/why you might want to do it, plenty of concrete examples, and a big, gorgeous pile of code to take home, which you can re-run at your leisure or just set up a giant Bayesian model code swimming pool in your mansion, a la Scrooge McDuck:

Slide1

For details, including the sign-up form: http://case.uoregon.edu/node/7

 

End of Term Reminder

Just a friendly reminder, since we’re approaching the end of the term. If you’re taking Bayes Club for credit, you need to:

1) attend Bayes Club (we’ve been keeping track of attendance, so you’re covered there)

2) send a one-page-ish write-up to Sanjay (sanjay@uoregon.edu) by the end of the term, describing what you’ve done and/or what you’ve learned in Bayes Club this term.

If you are signed up for credit and you think you’ll have trouble meeting one or both of these requirements, email Sanjay.

Bayes by hand, the sequel!

See the original post on doing Bayes by hand for some information about what conjugate priors are and why we might care. The demonstration today is showing that the beta distribution is a conjugate prior for a binomial likelihood function. Have fun!

Rose’s notes: https://www.dropbox.com/s/11ls35g9tpisfnf/beta%20binomial%20by%20hand.pdf?dl=0

R code: https://www.dropbox.com/s/j5wdmpgmdmjt9t0/conjugate%20priors.R?dl=0

desmos plot: https://www.desmos.com/calculator/etqxwwbbsg

 

Bayes Club Agenda

Hi all,

For our first meeting this term (today, 3:00-4:20 in Franklin 271A), we’ll go over the following topics:

1. Quick-n-dirty intro to Bayes: A refresher for those of us who haven’t been thinking about this since Spring, and a baptism by fire for those of us who are brand new. 😉 We’ll touch on some basic vocab and concepts, including possibly using this demonstration: http://setosa.io/conditional/

2. Planning this term’s schedule: The goal is to get a list of mini presentations lined up (“nuggets”, for those of you familiar with this practice in R Club). We’d love everyone to sign up for at least one little informal presentation. You can plan to talk about an article, a question or a problem you’ve run into, or you can plan to share something you’ve learned or figured out that you think would be of interest to the group. It can be long and complicated, or very quick and basic. The idea is to give each of us an excuse to do a little prep/thinking/reading/futzing with Bayes stuff outside of Bayes club at least once over the course of the term. The best way to learn is to do, after all!

3. Today’s nugget: Bayes by hand! Yep, over the summer, I (Rose) learned how to actually do Bayesian analysis by hand, and with real distributions (not just toy examples with coin flips). It’s a mess, but it was super helpful for me in terms of building my understanding of how Bayes’ theorem actually works in a practical sense, so I figured I would try it as a demonstration for Bayes Club. Bring a pencil, and brace yourself.

Bayes Club starts next week!

The first meeting of Bayes Club Fall 2014 will be Tuesday 10/7, 3:00-4:20pm in Franklin 271A. We hope to see you there!

To whet your appetite, consider reading this recent NYT article: http://www.nytimes.com/2014/09/30/science/the-odds-continually-updated.html It’s about how Bayesian inference is all hot in stats right now, and everyone’s all like, “Ooo you can do Bayesy stuff? You’re soooo cool!” It also gives a nice, very general-audience description of the difference between frequentist and Bayesian stats (with a coin-flip example, of course. What is it with the coin flips?).

For another excellent piece of broad-audience reading, check out this article in the Guardian: http://www.theguardian.com/science/life-and-physics/2014/sep/28/belief-bias-and-bayes?CMP=twt_gu It gives my favorite (to date) common sense explanation of Bayes’ Theorem – and it does so without referencing a coin!

If you’re still trying to decide whether or not you want to officially register for Bayes Club (vs. just showing up casually), check out our recent post on registration.