Joachim Vandekerckhove 🏖️

New mantra to time proper hand washing

Do you have a carbon monoxide detector?

The third class in the sequence goes more in depth on various common models

Thanks! It's all pretty high level -- "nonlinear models" is basically just logistic regression, because it's nontrivial but you can explain it from first principles

1 Course structure The following schedule gives the topics for each week. The schedule may be updated and fine-tuned as the quarter progresses. Any important changes will be notified well in advance. 1.1 Week 1: Formal logic • Formal logic and abstraction • Validity and soundness • Logical connectives and truth tables • More logical connectives, operator precedence, and basic propositional arguments • Reading: Navarro (2018a) 1.2 Week 2: Probability theory—the logic of science • Recap of truth tables • The rules of probability • Probability is the logic of science • From language to logic to probability • Group operators • Reading: Navarro (2018b) 1.3 Week 3: Distributions and the likelihood • Sample spaces and the probability mass function • Counts and combinations • The binomial distribution • Testing theories • Statistical summaries and statistical expectations • Reading: Etz & Vandekerckhove (2018) 1.4 Week 4: Applied probability theory, Evidence • Probabilities are areas • The cumulative distribution function • Conditional distributions • Bayes’ theorem • Reading: Etz & Vandekerckhove (2018) 1.5 Week 5: Revision and midterm • Revision of Week 1-4 • Friday: Midterm 1.6 Week 6: Linear models • The likelihood question • Evidence and plausibility • Linear regression • Reading: Platt (1964)

1.7 Week 7: Inference vs. estimation • Manual muscle testing study • Inference is moving plausibility around • Today’s posterior is tomorrow’s prior • Evaluating the prior; Cromwell’s Rule • Inference vs. estimation; Custom model construction • Reading: Etz, Haaf, Rouder, & Vandekerckhove (2018) 1.8 Week 8: Nonlinear models • Integration in R • Custom model construction • Linear models • Nonlinear models • Reading: Etz, Haaf, Rouder, & Vandekerckhove (2018) 1.9 Week 9: Inference in difficult situations • Interpreting the Bayes factor • M-closed, M-open, M-complete • The p-value • Reading: Wasserstein & Lazar (2016) 1.10 Week 10: Catch-up and revision • Fixed and random effects • Crossed random effects • Item response theory • Reading: Lecture notes

Class notes, accompanied by recorded lectures on each topic, comprise the required reading. Additionally, we provide a list of introductory texts that cover the material taught in this course. All readings will be made available as PDF files on the course website. • Etz, A., & Vandekerckhove, J. (2018). Introduction to Bayesian inference for psychology. Psychonomic Bulletin & Review, 25, 5-34. • Etz, A., Haaf, J. M., Rouder, J. N., & Vandekerckhove, J. (2018). Bayesian inference and testing any hypothesis you can specify. Advances in Methods and Practices in Psychological Science, 1, 281-295. • Navarro, D. (2018a). Learning statistics with R (version 0.6), Part II. • Navarro, D. (2018b). Learning statistics with R (version 0.6), Part IV. • Platt, J. R. (1964). Strong inference. Science, 146, 347-353. • Wasserstein, R. L., & Lazar, N. A. (2016). The ASA’s statement on p-values: context, process, and purpose. The American Statistician, 70, 129-133.

Sure -- slightly dated despite the date marker at the top, but this is the aspirational content. I usually don't get around to all of Week 10. The only prereq for this class is "Exploratory Data Analysis with R," which @jeffrouder.bsky.social usually teaches. The readings are mostly optional.

🙋‍♂️

Have been for years. Two weeks on propositional logic, three weeks on probability theory ("the logic of science"), four weeks on Bayes, and one bonus (i.e., content not included in the exam) lecture at the very end on the statistics wars, with a brief historical overview that includes p-values.

Chaser

Shot

Theory and practice of Bayesian inference using JASP YouTube video by OPAM Conference

Awesome YouTube tutorial by Johnny van Doorn, "Theory and Practice of Bayesian Inference Using JASP". Link to the video: www.youtube.com/watch?v=VhIm...

Blogpost:

www.bayesianspectacles.org/youtube-lect...

It's hard to be the bard

Gianmarco Tamberi and Mutaz Essa Barshim, right at the moment when they agree to share the gold after the men's high jump at the Tokyo 2020 Olympic Games.

Hard to capture in a still but Tamberi's face in this moment goes through so many emotions 😊

If the 95%CI does not contain zero, you can reject the null!
Of course, if the 95%CI does not contain zero, you can also reject the alternative. And you can do both of those things if it does contain zero. The world is your oyster!

Heh, I don't -- if it's not interactive I might as well be a traffic dummy and a tape recorder, when I could be playing Civ instead 😀

I should buy that book!

Should also point out that it's quite easy for Bayesians to divide out the prior for communication purposes -- at least if the conclusion at the end is about a discrete set of models/theories. Of course how you define a model is also subjective, but that's true for everyone

💯

Bespoke modeling plus the benefit of valid inference at the end. What's not to love!

Granted you can do bespoke modeling and compute a p-value at the end, but the user base for that seems small.

I was similarly a little surprised to learn that I enjoy teaching, specifically when it's 1-on-1 or small groups. We sometimes call that "retail teaching," as opposed to "wholesale teaching," which consists of lecturing to large amorphous groups and isn't really teaching at all, more like performing

"Arguendo, the data..."

I remember reading this and thinking that it's also indistinguishable from regular data collection. Disheartening to see the author did not draw this deeper connection.

Mom, come pick me up, the best lack all conviction and the worst are full of passionate intensity.

*sad jags noises*

Banner reads: STOP KILLING PEOPLE YOU FUCKING TWATS

"Hickok’s book is essential reading for both novices and hardened experts in the fields of cognitive neuroscience and the neuropsychology of speech and language." -- Biolinguistics

Resurrections and Insurrections in the Neurobiology of Language | Biolinguistics doi.org/10.5964/biol...

Lemme send you something in email when I get back to my desk 🙂

(I ask because the CrossRef MCP has yet to fail me at all.)

With an MCP or via the cursor chat itself?

How were you accessing these APIs?

Only two days left to submit your abstract for this year's @mathpsych.org meeting. Some things to consider:
1. The conference is in Canada.
2. If you cannot attend in person, there is a virtual track. Consider submitting there! You can pick the track during the submission process.
See you there!

PINGU: well now I am not doing it

me when I'm told to do literally anything by literally anyone

I worry we misinterpret the demand for numbers and graphs for a demand for expert insight. I don't know how many employers are interested in the latter

No cheating your last saved picture of a celebrity is your therapist.

(I feel like I win this one)