November 2015


Posted the blog. Good response today! I didn’t think it would be very widely read, but it’s getting quite a lot of readers.

Reading an old example from Lindley and Phillips, and they might have given me some new material for my talk this thursday. A good way of bringing out the importance of the sample space for classic stats.


Today was a good writing day. I got a lot on paper and feel like my draft of the super secret Bayes paper is almost done. Then just to send to EJ and have him rip it apart 🙂

The second day of the blog post had an almost equally good reception as day one! Wow. Seems my blog has a niche, because I’m pretty sure that nobody else is writing about this stuff.

Booking tickets to Germany and prepping for the talk. Luckily I’ve basically been giving mini-talks as blog posts for the past 6 months.

I’ve also been reading this old post from Mayo’s blog. Not sure what I think, I’ve never really read de Finetti. Maybe I should.


I’ve slipped in keeping up with this diary. So bad of me to miss almost an entire week. A lot of things happened this week. I finished the draft of the super secret Bayes project, went to Germany to talk with JP (and give my first talk), went to Rotterdam to meet with Daniel and give him Royall’s book, shared my slides from my talk (that were surprisingly popular), shared this excerpt, a nice set of tips for writing, Felix posted a neat new shiny app for PPV and FDR, Marsman and EJ submitted this interesting preprint, and a lot lot lot more. I have a renewed diary spirit! Onwards!


God hates all p-values. Even posterior predictive p values.

I’ve just seen this post from Gallistel, the last one in his Bayes for beginners series.

I also took another look at Gallistel’s old paper, “the importance of proving the null” and I just love this quote: “We are tempted to be vague to avoid the embarrassment of being wrong, but vague hypotheses can only be vaguely right.”

Uri Simonsohn is clearly on the side against the “null is never true”-ers. Seems we are on the same side for once.

Neuroskeptic has a new blog with a crazy funnel plot. He calls it an Avalanche plot. So so so crazy.


I love this quote from Christian Robert: “I will not be convinced that a method outperforms another method by simply looking at a series of simulation experiments.” It so succinctly sums up my views towards some of these simulation studies I see all over the place. The rest of the post is interesting as well, but that is the gem.


Leaving Amsterdam tomorrow. Writing some more today!

Richard has posted another “Neyman does science” blog, and it’s funny to see that even Neyman can’t resist the allure of evidence. Even the arch-behaviorist can’t avoid slipping into evidentiary language.


Travel day.

Daniel has a new blog, and apparently had his open access article taken and put up for sale on Amazon for 5 dollars. WTF? He did something very clever in response: Undercut with his own copy for 1 dollar.

Joachim got some media coverage for his publication bias work! Very nice.

Brendon Brewer has a great piece on the difference between Bayes and frequentists. I shared it on twitter and it got a lot of retweets! Other people apparently liked it as well.


Home again. Woke up at 4am due to jetlag :/ 😦 Not so great. But that gives me a chance to do some writing in the early hours of the morning.

Got my newest book: Jeffreys’s first edition of Theory of Probability! Super neat. It’s a way way bigger book than I expected. The pages are like 5 times thicker than books today.

There is a really cool group out there called “Statistics without borders” that gives free stat consultations to charities. A really really cool idea.

And apparently I am going to Psychonomics this year! I did not think there was any chance I’d be there but it seems I have a very generous benefactor. Looking forward to meeting lots of people and reuniting with others!


Great blog from Dorothy Bishop, “Who is afraid of open data?” She provides a list of six reasons people might not want to share their data, and I think 4 is one of the big ones. People just don’t want to be shown to have made an error. And yeah, it’s scary to think that someone might look into your paper and find some error that needs correcting. But that’s good!

Also writing this morning. Finishing up the super secret bayes project over the next few days I think.

Okay so this month’s diary died. I just lost track of it so many times with the travels that I decided to give up on it. Cut the losses and start again fresh with December. Live and learn.