Did you miss any of my writing last year? In addition to my big chapters I also released a few shorter pieces on the exploded logit model, symmetric prior models for positive spaces, and more.

A full pizza has fallen in the street and is constantly being run over. This has to be a metaphor for something.

A full pizza has fallen in the street and is constantly being run over. This has to be a metaphor for something.

Should initial conditions be forgot, and never brought to means? Should initial conditions be forgot, for con ver hence. For con ver gence my dear, for con ver gence, we’ll drink a coffee while we wait for con ver gence.

I’ve been looking into the details of a software called “Linger” and have not been able to get the Cranberries out of my head for days.

The horrors persist, but so does my writing. I reviewed all of my writing this year over on Patreon, www.patreon.com/posts/146964..., including a few pieces that until now have been Patreon exclusives.

'In particular, we firmly recommend: "Always write down the joint probability over all the settings you are interested in."' Many are saying!

The horrors persist, but so does my writing. I reviewed all of my writing this year over on Patreon, www.patreon.com/posts/146964..., including a few pieces that until now have been Patreon exclusives.

Naive averages failing once again.

I was looking into Voronoi diagrams for a side project, but I just could not follow the details. So I do want I am wont to do and dove into the details, working out a lot of geometry and making too many figures. HTML: betanalpha.github.io/assets/chapt... PDF: betanalpha.github.io/assets/chapt...

I was looking into Voronoi diagrams for a side project, but I just could not follow the details. So I do want I am wont to do and dove into the details, working out a lot of geometry and making too many figures. HTML: betanalpha.github.io/assets/chapt... PDF: betanalpha.github.io/assets/chapt...

Round and round it goes.

Round and round it goes.

If you incorporate the differences from weekdays and weekends as well (using a Dirichlet prior) you get down to 16 for a 50-50 chance: seehttps://drive.google.com/file/d/1ShvK_gHZOwOmtrf7Yy2ogi38YQaQJz_O/view.

Uncertainty quantification in business.

If you find yourself in need of distraction during the holiday season then consider dropping by the narratively generative modeling Discord server, www.patreon.com/posts/genera..., where we're discussing all aspects of modeling, implementation, and deployment.

Consumer PSA: A lot of stores will advertise an Xmas tree, but the package actually *contains* a Ymas tree. DON'T BE FOOLED.

Just sayin', turtle doves are gonna have to pick a side when the shit goes down between turtles and doves.

If anyone regrets not being able to attend my recent regression modeling course you can still access the slides, exercises, and recordings with a late donation. For donation instructions see betanalpha.github.io/courses/.

If anyone regrets not being able to attend my recent regression modeling course you can still access the slides, exercises, and recordings with a late donation. For donation instructions see betanalpha.github.io/courses/.

Shattered possibilities.

Shattered possibilities.

🪢 it’s too late; topologize 🪢

Exotic R^4.

Update from Tumblr: Wednesday! https://temporalecology.tumblr.com/post/802756052302135296

Anybody experienced with red-black binary search trees and open to a few questions, or know anyone who is?

Anybody experienced with red-black binary search trees and open to a few questions, or know anyone who is?

I'm not going to step into the recent zeitgeist of bad visualization takes, but I will remind everyone that the absolute best reference on good plot design (and so much more) is still www.principiae.be/X0100.php.

5 dimensions being high-dimensional, with intuitions from 1 and 2 dimensional spaces utterly failing, is a pretty good rule of thumb.

All models are great if you don’t check assumptions.

In 1980, Intel announced the 8087 Math Coprocessor, a chip that made floating-point 100 times faster. I opened up the chip, took photos of the silicon structures, and analyzed its circuitry. It's a very complex chip for its time. Let's take a look inside...

Data science research opportunity open to undergraduate students across all of Canada. I’ve got a proposal in there: “Classifying fish species with sound” 🐟🔊 A few summers ago I worked with a great quantitative ecology student from UBC so it’s very open to students from various backgrounds!

5 dimensions being high-dimensional, with intuitions from 1 and 2 dimensional spaces utterly failing, is a pretty good rule of thumb.

All models are great if you don’t check assumptions.

Had weird dreams last night and the only vivid detail I can remember is referring to a one-man band as "Music Man X", presumably because with all of those instruments the performer looks like a robot. Upon woken reflection, a one-man band rocking the Storm Eagle theme would be pretty sick.

My thesis analysis was never going to be published, but that's not the fault of the ol' machine. I almost feel nostalgia about the shower maximum detector. Almost. 🫡

Only a few more days to register for my charity regression course on Wednesday. All material, including slides and recordings, will be made available for those who cannot attend live. A few sponsored registrations still available. Registration details at betanalpha.github.io/courses/.

Only a few more days to register for my charity regression course on Wednesday. All material, including slides and recordings, will be made available for those who cannot attend live. A few sponsored registrations still available. Registration details at betanalpha.github.io/courses/.

I was today years old when I realized that “integration by darts” is a pun in “integration by parts”.

So much statistics discourse orbits around Bayes verses frequentist debates, but in practice Bayes verses frequentist is often a false dichotomy. Just as important as the inferential strategy is the underlying model, and how that model is specified. 🧵

VOD is up now. We ended up going for two and a half hours, talking about time series, hierarchical, and even fact or models in constrained spaces, what makes for a well-defined model comparison and what makes for a useful one, and more.

Going live in a few minutes!

understand me

So much statistics discourse orbits around Bayes verses frequentist debates, but in practice Bayes verses frequentist is often a false dichotomy. Just as important as the inferential strategy is the underlying model, and how that model is specified. 🧵

Academia would be fixed if people got entrance music at seminar and conference talks.

One week left to sign up for my online regression course raising funds for World Central Kitchen and United Farm Workers! Come learn how to actually justify statements like “a linear model is good enough for this application”. For course and registration details see betanalpha.github.io/courses/.

One week left to sign up for my online regression course raising funds for World Central Kitchen and United Farm Workers! Come learn how to actually justify statements like “a linear model is good enough for this application”. For course and registration details see betanalpha.github.io/courses/.

Friendly reminder that I’ll be hosting statistics office hours on Wednesday, www.patreon.com/posts/holida..., with the VOD available immediately afterwards if you can’t make it live. Amongst other topics a theme of dynamical models on constrained spaces seems to have emerged...

Isn’t it fucked that the words Vaseline and baseline are only one letter off from each other