All the material for my Bayesian Data Analysis course is available online, including the lectures, which we re-recorded this fall (some of them by @aloctavodia.bsky.social and Noa Kallioinen while I was on vacation). The video links are listed in the schedule at avehtari.github.io/BDA_course_A...
Huge thanks to @posit.co and hosts @mchow.com and @wesmckinney.com for having me on The Test Set!!!
posit.co/thetestset/e...
My favorite part was learning that I've been pronouncing "pandas" correctly all along. I've never felt more vindicated.
And y'all are wrong about candy corn. 😜
My @upshot.nytimes.com column on the dynamics that led to the sudden fall of Bill Cosby seems very relevant to a certain Harvard economist right now www.nytimes.com/2014/11/21/u... The Harvard Crimson article has become a focal point www.thecrimson.com/article/2025...
"“The cozy friendship between Epstein and Summers on display in the emails is disgusting and disgraceful,” Statistics professor Joseph K. Blitzstein, who teaches Harvard’s largest introductory statistics course, wrote in a statement to The Crimson."
Proud to have been JB's student.
I'm glad it was an impactful experience. I remember having you in the class and hearing about your music research!
Awwww thank you, I am very proud to have had you and @mayasen.bsky.social as students too! As for this quote, I just said what should be obvious to anyone with a decent sense of morality who read the emails, but sometimes it's important to say obvious things on the record.
Harvard professors responded with outrage to a tranche of emails showing a close yearslong correspondence between former Harvard President Lawrence H. Summers and sex offender Jeffrey E. Epstein.
Shawn A. Boehmer and Veronica H. Paulus report.
www.thecrimson.com/article/2025...
As always, nice reporting from the Crimson here
While Summers has broken no laws, and he’s free to correspond with whom he chooses, there’s a lot here to support the idea that he should not be taken as an authoritative spokesman on so many important issues
www.thecrimson.com/article/2025...
This is the stuff of authoritarian regimes. There isn't even the pretense that this is anything but persecution, unbound by law.
Yep, only possible for fall semester courses
Giving my Stat 110 final today, on Taylor Swift’s birthday. Last year the final was also on Taylor Swift’s birthday! So I had a homework problem earlier in the semester on the probability of this kind of coincidence.
@hadleywickham.bsky.social says at adv-r.hadley.nz/oo.html "Generally in R, functional programming is much more important than object-oriented programming... Nevertheless, there are important reasons to learn each of the three systems". Definitely both are valuable.
Thanks! :) Yes I did coin that (as far as I know), by analogy with object-oriented programming. OOP seems to have gone out of fashion but certainly still has some useful concepts such as encapsulation and inheritance, which are useful both for coding and for probability and statistics.
We collected lecture notes and blog posts by group members about recent topics in deep learning theory here. Hope it is useful!
pehlevan.seas.harvard.edu/resources-0
Free Weekly Econometrics Office Hours Email: davidbchilders@gmail.com or Sign up form: https://forms.gle/yVz8RtVALDXTmv5u7 Time: Wednesdays 10:00-12:00AM Eastern US (or by appointment) Location: Zoom Link https://bowdoin.zoom.us/j/7322488068 Who: Anyone. Grad students, researchers, government workers. Private sector is okay but in that case if your question requires work that exceeds the allotted time I may request to negotiate a consulting fee. What I can probably help with: Theory questions. Research design. Modeling. Particular expertise: Time series. Causal inference. Bayes. Structural approaches. Machine learning. Theory: Asymptotics. Statistical learning. Bayes/MCMC. Identification. Decision theory. Semiparametrics. Fields: I know most about macro (DSGE, heterogeneous agents, VARs, etc), but can follow along in applied micro (labor, development, health, etc) & some finance. Code: I think in R, can write Julia, and can get by in Python. I am likely to suggest you build a model in Stan. I know Stata but if it’s relevant to your question I suspect you can get better help elsewhere.
Since I am suddenly inundated with statistics-oriented followers, I should post again that I am still conducting free weekly open office-hours for anyone in the world with Econometrics questions, Wed 10-12AM Eastern or by appointment.
Details and sign up at donskerclass.github.io/OfficeHours....
It's amazing that a free chess app running on an iPhone can play at a 3600 ELO rating level. No human has ever gotten a rating of 2900+. The app is called Smallfish, which runs stockfishchess.org
The leading chess engines are now so much stronger than any human that this isn't really the case anymore. For a while there were some interesting "advanced chess" or "centaur chess" tournaments where a human worked with a computer, but I haven't seen that recently: en.wikipedia.org/wiki/Advance...
I love the Feynman quote that starts the article "Differentiating Under The Integral Sign" by Keith Conrad kconrad.math.uconn.edu/blurbs/analy...
It's also one of my favorite problem-solving methods. I thought it deserved a shorter name so I call it DUThIS (read as "do this")!
It would also be helpful for probability and statistics if the Riemann-Stieltjes integral were taught more often. For example, then we can write E(X) as the integral of x dF(x), where F is the CDF of X, rather than having to write separate formulas for discrete X and continuous X.
so do I, for over 15 years!
hey i know her
I made one for stats papers
concert poster
Yesterday was not only my first day on Bluesky, but also my first time acting since elementary school. I was in a Harvard Pops Orchestra concert playing two roles: myself and the Unconscious Statistician (of LOTUS fame). The musical performances were amazing (thankfully they didn't ask me to sing).