keep the werewolves at Bayes…
Following a light post by Rasmus Bååth on choosing a mascot for Bayesian statistics (even though the comments did not remain that light!), and suggesting puppies because of John Kruschke’s book, I commented that werewolves would be more appropriate! (Just to be perfectly clear, I do not think Bayesian analysis needs a mascot. And even object to the use of Reverend Bayes’ somewhat controversial picture.) To which Rasmus retaliated by sending me a customised cup with John Kruschke’s book‘s cover. Rising to the challenge, I designed my own cup by Googling a werewolf image and adding the text “keep the werewolves at Bayes… by sticking to your prior beliefs“. In retrospect, I should have instead used “to keep the werewolves at Bayes… run, little Markov chain, run!“. Maybe Definitely another cup in the making!
February 7, 2014 at 6:11 pm
I counter your Bayesian mug post with another mug post (+ an image of my cat):
http://www.sumsar.net/blog/2014/02/bayesian-mugs-galore/
Thanks again for the mug :)
February 7, 2014 at 10:13 pm
And the cat logically favours the werewolf. I am amazed at the collection of Bayesian mugs already available. And all of them turning to werewolves on full mean!
February 3, 2014 at 11:30 pm
A truly awsome mug that I believe will help with the most important task of the statistician: To strike fear in the heart of his collegues! (“Are you sure that your data agrees with the assumptions of the t-test? Perhaps a hierarchical Bayesian model would be more suitable? Oh it’s easy to code up in c++ . Or Fortran…”)
February 2, 2014 at 2:28 am
Do you know/play the RPG game Werewolf: The Apocalypse?
February 2, 2014 at 2:41 pm
Not at all! I do not play any game, as my son plays enough for four…!
February 2, 2014 at 1:45 am
Wow. That comment thread. I never thought I’d see people calling each other names over measure theory! I do love the internet :p
February 2, 2014 at 2:47 pm
wow indeed. it still remains a mystery as to why people get so intense on discussion forums. Do they truly think they could eventually convince the other party?
February 2, 2014 at 4:33 pm
It’s part a problem of form. In text it’s difficult to fix bad word choice or a poorly executed argument, feedback isn’t always instantaneous, and it’s easy to accidentally give offence. There’s a nice bit in there where Mayo mentions that she hasn’t gotten the hang of the tone of twitter yet, which I think is universally true for people and blog comments. At best, you’re trying to hit a rapidly moving target.
February 2, 2014 at 4:40 pm
Incidentally, I did have one relatively intense discussion on Gelman’s blog about cross validation (of all things), that was actually useful in helping me work out what I actually thought about it.
But I have this weird thing where I really like finding out I was wrong about that sort of thing (or on the wrong track, or just a bit off or whatever). I’m deeply suspicious of the utility of ideas that are easy enough for me to understand completely.
February 3, 2014 at 9:27 pm
“Do they truly think they could eventually convince the other party?”
Of course not. I’m well aware 99% of academia is a sham/scam and there’s nothing I can do about it. It still sticks in my craw sometimes though.