Sean Pinkney
LightNews LightNews
banner
spinkney.bsky.social
Sean Pinkney
@spinkney.bsky.social
940 followers 590 following 89 posts
I mostly post about probabilistic programming stuff, statistics, and R/Julia/Python (in that order). I'm a volunteer Stan developer and citizen scientist (papers on arxiv). Currently my day job is doing marketing analytics.
Posts Replies Media Videos
spinkney.bsky.social
Halloween @mc-stan.org
October 29, 2025 at 9:22 PM
spinkney.bsky.social
Stan 2.37 will be out sometime next week. I helped get the updated simplex parameterization and sum-to-zero matrix (both rows and columns sum to zero).

We found the new simplex parameterization to be much better in the extremes of the simplex and across very small alpha values of the dirichlet 1/
June 4, 2025 at 12:45 PM
spinkney.bsky.social
Ask me any questions you may have about this! #stats
learnbayesstats.bsky.social Pierre-Simon Laplace @learnbayesstats.bsky.social · May 30
🎙️ Ep. 133 is out now!

@alex-andorra.bsky.social chats with ‪ @spinkney.bsky.social
& Adrian Seyboldt about making Bayesian models more efficient without losing rigor — zero-sum constraints, Cholesky tricks, practical wins & more

🎧 learnbayesstats.com/episode/133-...

#Bayesianstats #podcast #LBS
Learning Bayesian Statistics – Laplace to be for new & veteran Bayesians alike!
Laplace to be for new & veteran Bayesians alike!
learnbayesstats.com
June 2, 2025 at 1:23 PM
spinkney.bsky.social
I've come on here after months away to let everyone know that I apparently really like trance music
May 7, 2025 at 1:12 PM
spinkney.bsky.social
Are there any other statistics nerds in the NoVA area that want to get together?
March 27, 2025 at 11:05 AM
spinkney.bsky.social
Promising results from the Pymc folks! #bayesian #pymc #stats #stan

discourse.mc-stan.org/t/updated-ch...
Updated cholesky corr parameterization testing
Promising results from the Pymc folks Confirmation that the det Jacobian is correct and early tests show improved sampling!
discourse.mc-stan.org
February 28, 2025 at 7:39 PM
spinkney.bsky.social
Fellow probabilistic programmers I came up with a parameterization for a Cholesky factor of correlation matrices that looks good. I'm asking to test and see if it works well for you discourse.mc-stan.org/t/updated-ch...

#rstats #statistics #bayesian
Updated cholesky corr parameterization testing
Hi everyone, I’m asking if people could test out this parameterization of the Cholesky factor of correlation matrices? From my tests on a macpro m1, it looks to be much faster and I can easily sample...
discourse.mc-stan.org
February 20, 2025 at 1:50 PM
spinkney.bsky.social
I don't know wtf this is but I'm super into matrices so the non-matrix theology is a non-starter for me
February 17, 2025 at 6:08 PM
spinkney.bsky.social
I will only be impressed by you if you solve a long standing mathematical conjecture and win the Tour de France within 2 years of each other. Otherwise, what are you even doing?
February 3, 2025 at 2:30 PM
spinkney.bsky.social
Basically he Colin from What We Do in The Shadows
jasonkoebler.bsky.social Jason Koebler @jasonkoebler.bsky.social · Jan 29
some really great tips here:
“Insist on doing everything through ‘channels.’ Never permit short-cuts to be taken in order to expedite decisions.” “Make ‘speeches.’ Talk as frequently as possible and at great length. Illustrate your ‘points’ by long anecdotes and accounts of personal experiences. Never hesitate to make a few appropriate ‘patriotic’ comments.” “Bring up irrelevant issues as frequently as possible.” “Haggle over precise wordings of communications, minutes, resolutions.” “‘Misunderstand’ orders. Ask endless questions or engage in long correspondence about such orders. Quibble over them when you can.” “In making work assignments, always sign out the unimportant jobs first. See that the important jobs are assigned to inefficient workers of poor machines.” “To lower morale and with it, production, be pleasant to inefficient workers; give them undeserved promotions. Discriminate against efficient workers; complain unjustly about their work.”
January 31, 2025 at 2:08 AM
spinkney.bsky.social
"Painting to be stepped on." Do not step on the painting!
January 2, 2025 at 9:26 PM
spinkney.bsky.social
Tried zyn and that shit was so dysphoric. I don't smoke or take nicotine so probably too much. I had 1 slim pouch (2/4 strength) for 5 mins and then 10 mins later another pouch for 15 min. Initially I felt lightheaded, then I started sweating, felt nauseous, and couldn't wait for the effect to end
January 2, 2025 at 8:19 PM
spinkney.bsky.social
I have a 2017 MacBook pro and work supplied me with a m1max 32 GB about 3 years ago. I'm thinking I'll buy two or three mac minis and use them as a cluster. My son wants to start a twitch/yt stream and I think I can set up game screen capture better this way too. Any thoughts?
December 26, 2024 at 2:30 PM
spinkney.bsky.social
Public service announcement on zero-sum vectors

discourse.mc-stan.org/t/zero-sum-v...
Zero sum vector and normal distribution
Zero-sum vector and normal priors The release of 2.36.0 introduces a new constraint transform in the Stan language. This constraint has been discussed here and, I’m sure, in other topics over the year...
discourse.mc-stan.org
December 24, 2024 at 11:48 AM
spinkney.bsky.social
I'm going to have time to do 1-2 contributions for the summer 25 release. Here's my list to choose from, what is most interesting to you?

- adding lower/upper bounds to ordered vectors (removing positive ordered since it's achieved by lb=0)
December 2, 2024 at 11:46 AM
spinkney.bsky.social
A new release of Stan is coming discourse.mc-stan.org/t/cmdstan-st....

Highlights:

New constraints for stochastic matrices and zero-sum vectors

Easier user-defined constraints

Improved diagnostics

New distribution, beta_negative_binomial
CmdStan & Stan 2.36 release candidate
I am happy to announce that the latest release candidates of CmdStan and Stan are now available on Github! This release cycle brings new constrained types, improved diagnostics, and a new distributio...
discourse.mc-stan.org
December 2, 2024 at 11:26 AM
Reposted by Sean Pinkney
coatless.bsky.social
Woah, the new #Stan Playground lets you run MCMC in your browser & analyze the chains using #WASM powered #Rstats & #Python! 🎉

✨ No installation needed - just write, compile, sample & analyze
🎯 Perfect for teaching & collaboration (share links ftw!)

🔗 Try it: stan-playground.flatironinstitute.org
Stan Playground
stan-playground.flatironinstitute.org
October 29, 2024 at 3:37 AM
Reposted by Sean Pinkney
avehtari.bsky.social
Call for StanCon 2025+ discourse.mc-stan.org/t/call-for-s...

StanCons have been the best conferences where I ever have been and suitable also for non-Stan people

#Bayesian #Stan #StanCon
Call for StanCon 2025+
We are looking for volunteers to organize the next StanCons for 2025 (and beyond!) If you are interested in making a proposal (or even just discussing the possibility of making one), please consider ...
discourse.mc-stan.org
November 12, 2024 at 4:17 PM
spinkney.bsky.social
First citation for my flexible parameterization of Cholesky factors of correlation matrices (arxiv.org/abs/2405.07286) this is in "Asymptotics for parametric martingale posteriors" by Edwin Fong, Andrew Yiu paper link at arxiv.org/abs/2410.17692.
November 29, 2024 at 5:31 PM
spinkney.bsky.social
Easy block list
stevecicala.bsky.social Steve Cicala @stevecicala.bsky.social · Nov 20
Have you subscribed to any block lists? There are some people doing the public service of compiling troll lists. Subscribe to the list, and those accounts cease to exist for you. Here's one, perhaps others have suggestions.

Block liberally and hopefully things clear up.

bsky.app/profile/numb...
November 21, 2024 at 1:05 PM
Reposted by Sean Pinkney
cscherrer.bsky.social
I got a Probabilistic Programming starter pack going. Hit me up if you're involved with #probprog R&D and want in!

go.bsky.app/JfvubEf
November 19, 2024 at 4:10 PM
Reposted by Sean Pinkney
josiah.writes.news
NEW: "Moving Off X: An Open Letter from UK Journalists"

UK political journalists: "Twitter was, for a long time, the place to be for UK politics. We believe that time is now over...

"We have taken the decision to move over to other platforms, in particular Bluesky" docs.google.com/document/d/1...
We have used Twitter – now X – for many years. It has been a resource for story-gathering, network-building, and a megaphone for getting public-interest stories to a wider audience. And it has been a well of ideas and discourse. Unfortunately that well is now poisoned. All of us have seen X become more toxic and broken since Elon Musk took over in 2022. Feeds have become less useful. Engagement has plummeted, except for those who will pay. Replies gain traction not through merit, but through the corrupted blue-tick system. We have seen hate speech and abuse deliberately dialled up and amplified, boosted not just by the algorithm, but by the owner himself. And England’s racist summer riots were one of many factors suggesting to us that X is not a neutral platform, but a space that now seems to glorify misinformation at a great cost to our public sphere. Millions of people are ditching X, and we are doing so, too. We have taken the view that it is no longer a useful tool for objective reporting, but a weapon being wielded by a narrowing ideological set. A tipping point is coming. Perhaps we are already in it. After much consideration, we have taken the decision to move over to other platforms, in particular Bluesky. For some of us that means instantly deleting our X accounts, for others it is a more gradual process. But we are all putting the majority of our efforts into building more constructive online spaces elsewhere for political dialogue and reporting. Bluesky is becoming a new hub for political discussion, and we encourage other UK journalists to join and become active. We are already seeing far higher engagement on our posts there than we’ve seen recently on X. And importantly, the mechanisms to limit abuse actually work. We enjoyed Twitter while it lasted. It played an important role in shaping many of our careers. And it was, for a long time, the place to be for UK politics. We believe that time is now over. Twitter is dead. Long live Twitter
November 15, 2024 at 11:51 AM
spinkney.bsky.social
Statisticians being based
October 28, 2024 at 10:39 AM
Reposted by Sean Pinkney
drewtoothpaste.bsky.social
finally!
not interested in "fame and fortune?" try "just fame no fortune"
October 27, 2024 at 2:52 PM