I know of a 40,000 patient clinical trial where the SAS programmer determined the key secondary outcome as positive if Y=time to stroke < maximum follow-up time. Y=missing -> positive outcome because with SAS missing < anything. The error still stands.

I feel like #lazyweb isn't what it used to be, but anyone in #econsky happen to know of a good package for implementing fuzzy regression discontinuity designs in python?

Does anyone want a survivorship bias shortbread

Wondering what we lose as tech writers write for AI training vs humans. The lovely `polars` docs often use nice reprexs to demonstrate syntax but then note if the technique shouldn't be used in such a trivial case. Would someone write this now or worry AI picks example w/out context?

Great taxonomy for managers, as well! So important to be clear when feedback is a must-have, nice-to-have, or curiosity question you don’t reasonably expect to get answered

Continuing my #python Rgonomics posts for the #rstats crowd, wrote a short post about the different ways to use (non-polars) user-defined functions in a pipe Not the most groundbreaking thing, but one of those "the thing I needed to read a bit ago" posts (1/) www.emilyriederer.com/post/py-rgo-...

Every year around this time I’m surprised/disappointed to find there aren’t good places online to print custom 365 day calendars Do you people like.. not maintain lists of 400+ pithy thoughts on the same topic? (Seriously, does anyone know where this can be done?)

1. This is reprehensible and should not exist. What are we doing? 2. Mildly amusing picturing AI me, learned from my digitaltraces; family and friends reaching out in their moment of need, and I'm just spouting off about data quality checks and column naming conventions

Favorite talks from day 1 of @causalscience.org 's #CDSM2025 🧵

Truly one of the best data science events that I look forward to all year long! No sales, no hype, just exquisite thoughtful yet practical methods. Taking two days off work to attend. Don't miss out!

A twinge of jealousy for the ecologist-types who are getting one heck of a natural experiment out of the FAA flight reductions Terrible if you’re trying to live in a functional society, but pretty lovely if you do causal analysis on pollution or emissions

Python SC accepted PEP 798 PEP: peps.python.org/pep-0798/ Acceptance: discuss.python.org/t/pep-798-un... So this: [*row for row in list_of_lists] Will do the same thing as this: [x for row in list_of_lists for x in row]

⏰ Only 1 week left to register for #CDSM2025! Join us Nov 12–13 for two days of talks & debates at the intersection of causality, data science & AI. 💻 Online | 🎟️ Free 👉 causalscience.org

My theory is that for practitioners, regression models should be like pocket money: you get a fixed number per week to do whatever you like with until we're sure you won't blow the whole lot on silly stuff, get caught up in a get-rich-quick scheme, or accidentally leave them in a drawer somewhere.

The way python and R foster inclusion directly contributes to their success: joyful places to exist, a steady flow of new maintainers, and a delightful collection of niche tools empowered by wildly different expertise coming together Watch the new python documentary for more on PSF’s work here

jacobtomlinson.dev/posts/2025/t... Highly relatable for anyone that has ever written a line of code used by other people Lovely little post from @jacobtomlinson.dev

Do you enjoy playing chartle.cc? We’re opening up to the community! Are you a scientist, journalist, researcher or just a data-enthusiast and would like to submit your own data (it should be a country-based time series)? Get in touch with us - we’d love to feature it!

I have no idea whether or not this is a good paper, but where the abstract cut off is the ultimate cliff hanger

It occurs to me that the CFPB epitomizes what Abundance calls for: Legitimacy through results, not process. They got a lot done, and that is precisely why conservatives put a target on their back. Yet, Klein and the other abundance bros never talk about them.

I have the privilege of working with fed agency folks. They've been jerked around, forced to move, had offices taken, lost resources, put on leave, fired, re-hired, *shot at*, and now furloughed. Still they return each day to help make America a bit more informed, healthier, and safer.

Tomorrow I have to take 25 blurbs of text, copy paste them one by one into a UI, get an ID and link, and then go paste those into some docs I’m reminded how much more I value of a good REST API or RPA/headless browser (Playwright!) over AI for the kind of work I want to delegate to a robot

1. Bluesky doesn’t downrank links 2. You can choose your own algorithm from among thousands 3. The whole app is open source This keeps it a free and open platform by design, leading to better click-throughs and traffic to your site. We 💙 the open web.

Really insightful post from Julie Tibshirani (spotted in LinkedIn, can't find on Bsky) reflecting on #rstats 's unique governance structure and what can be learned for other languages jtibs.substack.com/p/if-all-the...

Thrilled to once again be able to bookmark posts I will never once revisit like a chipmunk hoarding nuts before winter 🐿️ The really killer feature that the former place never had would be to search or organize your bookmarks 🙏🏻

Amusing how 99% of people using LLMs forget how these things work: They are advanced probability machines. They generate the next most likely token (word) based in the input and their training. Under the hood, it’s a giant matrix multiplication that has eerily good output.

Tonight’s the night: PyData Seattle’s CFP & Labor Day Flash Sale will CLOSE tonight at midnight! 😱 This is your final chance to share your technical talks with our technical people, and even get 25% off your ticket! Don’t let this opportunity slip by, GO GO GO! pydata.org/seattle2025

Python documentary is a fantastic long weekend watch! Love the balance in the story between the tech and community 🍿

➡️ Deadline approaching—only one month left to send in your papers and presentation proposals for #CDSM2025! 🚨 𝗖𝗮𝗹𝗹 𝗳𝗼𝗿 𝗣𝗮𝗽𝗲𝗿𝘀: 𝗖𝗮𝘂𝘀𝗮𝗹 𝗗𝗮𝘁𝗮 𝗦𝗰𝗶𝗲𝗻𝗰𝗲 𝗠𝗲𝗲𝘁𝗶𝗻𝗴 𝟮𝟬𝟮𝟱 🚨 📅 𝗡𝗼𝘃 𝟭𝟮–𝟭𝟯, 𝟮𝟬𝟮𝟱 (𝗩𝗶𝗿𝘁𝘂𝗮𝗹) 📥 Submission Deadline: 𝗦𝗲𝗽𝘁 𝟯𝟬, 𝟮𝟬𝟮𝟱

The amount of alpha in this post is off the charts www.seangoedecke.com/good-system-...

Screenshot of cases from the U.S. CDC’s website showing the number of national measles cases indexed by rash onset date. Are they going down?

Dev and deploy require of diff skills. @getdbt.com bridges the gap for #data pipelines, but what about MLOps? I explored how @posit.co {orbital} + dbt delivers "good enough" zero-infra MLOps for batch model scoring in database from python or #rstats models www.emilyriederer.com/post/orbital... 1/

Sometimes I think I need a Quarto extension that explicitly removes any of my footnotes from the output to support my workflow of: - think of multi-paragraph tangent for anything I'm writing - satiate myself by sticking it in a footnote - save readers from sending them down my many tangents

A simple design framework that will massively improve your slides and your public speaking clauswilke.substack.com/p/slides-tha...

Very neat project/paper decomposing model AUC to within-cluster ranking and between-cluster calibration github.com/asudjianto-x... Model eval is hard! I've always thought clusters in ranking is esp interesting because it inflates utility for use within a subset of training population, e.g. 1/2

"What Happens When Politicians Meddle With Economic Data: Argentina’s Example" www.wsj.com/economy/trum...

Sen. John Cornyn, R-Texas, voted for cuts to the Consumer Financial Protection Bureau, but he’s also routed more than 800 of his constituents’ complaints to the agency — the most of any current lawmaker from either party, ProPublica found. By @joeljacobs.bsky.social

"PhD-level experts in your back pocket" is a completely nonsensical description of AI but a pretty good description of social media if you follow the right people

"And if you thought this was unimportant to you, remember BLS calculates the CPI used for Social Security COLAs, as well as the headline CPI used to calculate TIPS yields. Oh, also the inflation adjustments to the marginal tax rate thresholds…" #econsky econbrowser.com/archives/202...

Issue 7 of RDM Weekly is out! 📫 This week's issue includes: - Making Messy Data @nrennie.bsky.social - Data Analysis with AI @gaborbekes.bsky.social - Column Names as Contracts @emilyriederer.bsky.social - Resources for RCTs @j-pal.bsky.social and more! rdmweekly.substack.com/p/rdm-weekly...

Really enjoyable listen! Think you know testing? Think you know git? Still a thoughtful discussion of good workflows putting these together. Had not thought about git stash as they describe before

Terence Tao (@teorth.bsky.social) has written a thread on Mastodon about the impact of the federal grant freeze to UCLA, particularly to his own field of Mathematics. UCLA's IPAM (Institute of Pure and Applied Mathematics) could shut down entirely mathstodon.xyz/@tao/1149568...

Here’s a graph from the Economist showing GDP data from official government sources versus private estimates. Free countries, like the US, don’t manipulate econ data. Less free countries do. Trump is seeking to break that for short-term PR. (Yes I know GDP is BEA and not BLS.)

Time's almost up for getting your talk submission in for Data Behind The Scenes Conf! Essentially 2 days left to decide what kind of tea you want to spill about doing data work to an audience of other data folks to commiserate! #dataBS bit.ly/dataBSconf-cfs

A favorite part of my Chicago commute: walking past a building engraved “to safeguard wealth men established banks” that was built in… 1929 I’ve yet to find the complementary building that says “to safeguard banks men established the FDIC”

32. The bootstrap is basically magic

Only a few days left to apply for this normconf spiritual successor! Come tell your stories about *really* working with #data -- all nuts & bolts, triumphs & tribulations, no nonsense

Technical writing is hard bcs "writing is thinking" but we often should tell our story not in the order we worked. Solution? I wrote a quick post on how @quarto.org 's embed shortcodes can reframe technical writing as reproducible evidence curation www.emilyriederer.com/post/quarto-... 🧵 (1/n)

I have deleted the data that were in the database and which you were probably saving for production Forgive me I was panicked So artificial And so intelligent

Not to miss the forest for a tree on this thread but an underappreciated angle imho: Vibe coding tools that can delete an entire prod database are probably aren’t meticulously securing that database Don’t vibe code? Cool, but if any app you use does, it could still be your data privacy at risk

I'm intrigued by the potential* for @posit.co 's {orbital} package to translates ML prediction logic to SQL, but for now it doesn't *officially* support xgboost In this post, I explain some hacks to use orbital + xgboost and what I learned about ONNX www.emilyriederer.com/post/orbital... #data