Some lecture notes on spatial and temporal modelling, illustrated using R - darrenjw.github.io/spatio-tempo... - #rstats #rspatial #quarto

The Trump Accounts seem to be functionally the same as the the "Child Trust Fund" introduced by the UK Labour Government under then Chancellor Gordon Brown. Those funds ran from 2002-2011. I have not seen a comparison made, or if the UK experience was regarded as a success. Be interested to hear.

New paper on analytic filtering, prediction and smoothing for a class of exponential family models. Interested in comments. We continue to work on this abstract setup with more general weight functions and loss functions --- hope to report results soon. tinbergen.nl/discussion-p...

RIP, may her memory be a blessing #econsky marketdesigner.blogspot.com/2025/12/kate...

This Wednesday: hear pitches from Harvard researchers, including teams from the IQSS Startup Foundry, about the latest AI projects ready to become ventures and looking for partners. Register at the link for the Mind2Market Demo Day on Dec 5: www.grid.harvard.edu/mind-2-market

fair play from Gerko on.ft.com/4pJUely

Yale’s Institute for Foundations of Data Science (FDS) @yaledatascience.bsky.social is seeking applications for postdoctoral positions. These are cool, generously supported, competitive positions, expected to last 2-3 years, for independent scholars working on the foundations of data science.

"“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.

Portable mortgages are like your cell phone plan: when you upgrade the phone, the plan comes with you. Let your mortgage follow you to the next house. That unfreezes movers—especially downsizers—and unlocks inventory. Lemme explain why I love this idea:

Interesting time series experiments paper by Peng Ding and coauthor: arxiv.org/pdf/2510.22864 Connected to my 2018 JASA paper with Iavor Bojinov.

Very sad about Ron passing away. One of the most stimulating econometricians I have come across during my career. A very fine scientist, always pushing areas he was interested in to the highest standards, not following the crowd. A wonderful person to spend time with.

Congrats to my colleague Lucas Janson who has been promoted to a tenured Professorship at Harvard Statistics. He works on high-dimensional statistics and machine learning. You can read more about his remarkable career at lucasjanson.fas.harvard.edu

This seems very interesting. I would be keen to hear about experiences with Refine.

There are accusations of impropriety against Lisa Cook, a fine economist, once again. Last time (around her confirmation as a Fed Board Governor) the specifics were public and obviously nonsense. This time, they may be buried in lawfare for a while. I stand with Lisa D. Cook.

"Importantly, this extensive work would not have been possible without the support by multiple NIH grants now being subject to profound cuts, no less $2.6 billion specific to Harvard."

N. Gregory Mankiw, Robert M. Beren Professor of Economics at Harvard University, presented the 2025 Martin Feldstein Lecture on "The Fiscal Future." https://www.youtube.com/watch?v=B916t3od8fM

Fun paper on the Repo market www.federalreserve.gov/econres/note...

Excited to announce the #OxfordLLMs tuition-free AI workshop for social scientists! Join us for 5 days of lectures, coding tutorials, and collaborative research, to learn hot to use LLMs for social science🙂 Dates: 22-26 September Venue: @nuffieldcollege.bsky.social llmsforsocialscience.net

We've published the inaugural editorial for our new journal RSS: Data Science and Artificial Intelligence 📚 We are open for submissions! 🥳 academic.oup.com/rssdat/artic... #DataScience #AI

Tonight, #RES2025 will present the 2024 Denis Sargan Econometrics Prize to Rahul Singh & Liyang Sun for their #EctJ article on doubly robust analysis of local average treatment effects and more general complier parameters res.org.uk/2024-denis-s...

Shunichi Amari has been awarded the 40th (2025) Kyoto Prize in recognition of his pioneering research in the fields of artificial neural networks, machine learning, and information geometry www.riken.jp/pr/news/2025...

My Vital Statistics newsletter is launching in 6 weeks. It’s for anyone curious about data, models, and what they can teach us about the world. Sign up now: kareemcarr.substack.com

Congratulations to PhD student Rocco Caprio on the acceptance of his first paper, "Error bounds for particle gradient descent, and extensions of the log-Sobolev and Talagrand inequalities" arxiv.org/abs/2403.0..., on functional inequalities and algorithm convergence - it will appear in JMLR.

Mixed day. With TF team grading the Stat111 course I teach with Joe Blitzstein. Harvard is an extraordinary place to work. Grading is a JOY with them. Also talked to my tenure track Stats colleagues who were having their U.S. Govt grants cancelled today solely for working at Harvard.

Assume: 2% inflation, 2% real wage rises, 9% returns. An institution can spend 5% a year on employing the same number of folk each year for ever. If returns are taxed at 21%, net returns become 7.1%. Can then spend 3.1% a year. Tax reduces spend from 5% to 3.1%: about a 40% fall.

Ever wondered why some charts are more effective and engaging than others? I’ll be going through all the theory, evidence and principles behind compelling data storytelling from this Wednesday. Free to watch for anyone who is interested: www.eventbrite.co.uk/e/the-fundam...

I need the “Campaign in prose. Govern in econometrics.” t-shirt.

Congratulations to Grace Wahba for winning the 2025 International Prize in Statistics, in recognition of her ground-breaking work on smoothing splines, which has transformed modern data analysis and machine learning 👏👏👏 Read more: rss.org.uk/news-publica...

Admirable article by Harvard CS colleague Boaz Barak on "professionalism" in the classroom www.nytimes.com/2025/05/02/o...

Congratulations to my Harvard Statistics colleague Jun Liu for being elected to the (U.S.) National Academy of Sciences. Jun's work on Monte Carlo and sequential learning has been an inspiration for me. The breadth of his research is extraordinary. statistics.fas.harvard.edu/people/jun-s...

Congrats to our Harvard Statistics colleague Mark Sellke for being award the Rollo Davidson Prize in probability: msellke.com His research is extraordinarily wide ranging over probability, mathematical physics, theoretical computer science, algorithms and statistics

"I'm going to govern in econometrics" is a winning campaign slogan for the ages.

Congratulations to Chris Walker after his Ph.D. defense at Harvard Econ. You can read about his research on Bayesian econometrics at sites.google.com/view/cdwalke... He is to be an Assistant Professor of Econ at Duke in the fall.

I love the Economic Journal referees ❤️ Revised version of a paper received on April 15. Sent out to original reviewers on the same day. All four referees have come back to me in less than 10 days. Paper is now conditionally accepted, subject to checks by the EJ Data Editor 🚀 @resmedia.bsky.social

Nice panel on trustworthy AI with Donoho and Jordan, chaired by Tracy Ke. April 29th. www.niss.org/events/copss...

Congratulations to Stefanie Stantcheva (@s-stantcheva.bsky.social) of @harvard.edu, winner of the 2025 John Bates Clark Medal! #econsky www.aeaweb.org/about-aea/ho...

An expert in econometrics, Kevin Dano has joined the Princeton Economics Department as Assistant Professor of Economics! Having completed his Ph.D. at UC Berkeley, his work focuses on panel data, networks, and dynamic discrete choice models. Read more: economics.princeton.edu/news/econome...

Steve Kerr wore a Harvard basketball shirt after the Warriors’ win tonight: “Yes, this is me supporting Harvard. Way to go. Way to stand up to the bully.”

I hope you will read the letter from President Garber www.harvard.edu/president/ne... and particularly take up his request that you read what the U.S. Government demanded of Harvard.

Happy to receive an NSF grant on studying external validity in RCTs (joint with Emily and Arun). Many thanks to the excellent work of the program coordinators and reviewers. Crossing fingers for us all

I am profoundly saddened and alarmed by Columbia University and Paul Weiss law firm’s capitulation to the increasingly dictatorial Trump administration.

Can't wait to combine this with the Poisson pseudo-maximum likelihood high-dimension fixed effects. I would call it PLRDPPMLHDFE.

Modern "Statistics" (as opposed to probability theory) started roughly in C18 with the collection of demographic & econ data so a modern state could effectively function. We should all want high qual measurements of our society & economy, well explained and PUT IN CONTEXT for the public.

I do wish Dr Murray well. Joe Blitzstein @stat110.bsky.social and I have used some of her drawings in our Stat111 class at Harvard!

I served on FESAC and its predecessor.  Even earlier,  I worked at BEA on the national income and product accounts.

link 📈🤖 Statistical inference for Levy-driven graph supOU processes: From short- to long-memory in high-dimensional time series (Mehta, Veraart) This article introduces Levy-driven graph supOU processes, offering a parsimonious parametrisation for high-dimensional time-series, where dependencies

A touch of self-advertisement: I'll be giving a research talk next Tuesday at the Online Monte Carlo Seminar (sites.google.com/view/monte-c...) on one of my favourite topics: the convergence properties of the Random Walk Metropolis algorithm. Please join us!

link 📈🤖 On (in)consistency of M-estimators under contamination (Klooster, Nielsen) We consider robust location-scale estimators under contamination. We show that commonly used robust estimators such as the median and the Huber estimator are inconsistent under asymmetric contamination, while the T

I have a new paper on “Prediction Sets and Conformal Inference with Censored Outcomes” with the great Weiguang Liu and Elie Tamer. Prediction Sets and Conformal… what? I know, I know! Here comes a (very) short thread on what we do in the paper. (1/5)

In order to facilitate housing growth and infill development, South Bend offers 27 pre-approved housing plans, ranging from single-family homes to duplexes and six-plexes. They provide site and building plans and pre-approve housing within two days. More of this.