I did update the figure a bit and added labels, but it probably works better as sequential slides in a presentation. If you want the code and the pptx just drop me a dm and I’ll send it all your way.

We streamlined six new DID-like estimators and created this tutorial for implementation in R. yiqingxu.org/packages/fec... Hope you no longer need to spend months figuring out what these estimators are and how to use them.

Foods I like that have been deemed “old man desserts” by friends and family: Rice pudding Oatmeal raisin cookies Rum raisin ice cream Peanut brittle These are all underrated desserts!

Fuckin Tim Apple

Hit the point in my phone’s life cycle where Apple fucks with the battery life and (probably) pays my friends to start telling me how old it looks when i check it.

It's annoying that I have to be an expert at navigating all these food ordering apps otherwise I'll be paying 10-20% more for everything.

I say this like once a week but it’s worth appreciating regularly: we’re living in the golden age of food. The standard for quality, taste, and innovation is insanely high right now. This is both for restaurants and home cooking.

The unreasonable effectiveness of a line chart with time on the x-axis.

This 90s alt rock playlist is saving my life right now.

Hello everyone! I teach a class for graduate students on methods for reproducible research. 🔧 🔢 I would interested to hear about the barriers you face in implementing reproducible practices in your research (e.g., version control, code review, reproducible documents).

It’s happening!!

Mosaic v0.12 is out: database-powered scalable, interactive visualization! 📈 One new addition is support for dynamic changes in the backing data. Move between smaller and larger samples to balance speed and comprehensive coverage.

I once tried to convince the Motif engineering team to run our app on a bunch of Mac Minis we would put in a closet at my house.

New post up! ✨ Exploring AT Protocol with Python to visualize the #databs social graph! davidgasquez.com/exploring-at... Took less than 1 hour to get the data and plot it. Amazing what you can do with open APIs and great SDKs!

Other examples of things that the (rapidly growing) data geeks on Bluesky can help with: improving and accelerating graph analytics to improve the default Discover feed: bsky.app/profile/dana...

Reminder that the blog is here too! @statmodeling.bsky.social Please repost 🙏

ICYMI last a couple weeks ago, how we built the Motif query engine and language for analyzing event sequences: www.motifanalytics.com/posts/how-to... Think of it as an extension of Regular Expressions matching from strings of characters to sequences of events.

I was definitely not expecting this explanation.

I’ll know Bluesky has really made it when there are tons of spam accounts following and DMing me.

Hey data nerds! You have just a week and half left to submit a talk to Data Council '25 (in SF next year!). I'm running the analytics tracks. Any interesting ideas we should be covering on the analytics track? Any brilliant people that fly under the radar that we should elevate?

Awesome, just what I wanted, the sun rises at 6:30am now.

Looking for talks for the Data Science and Algorithms track I'm hosting at Data Council next year. It's in Oakland April 22-24th. You can submit your talk proposal here: www.datacouncil.ai/cfp-2025 It's a fun/friendly conference with an amazing lineup, hope to be reading your submission soon!

My Bluesky notifications are blowing up! The exodus is happening!

So how long can Twitter operate with dwindling ad revenue? Has anyone done a good write up of the different endgames that are possible?

New Simpson’s paradox example just dropped.

Naming things is hard when you're building an analytics product! Here's one: - column - dimension - variable - attribute - covariate - series - feature are all used to refer the same thing. There are differences in definitions but even picking the "correct" one can just confuse people.

🎉New Motif demo day! 🎉 loom.com/share/dda8f3... I'm proud of our new icicle visualization which seamlessly toggles between disaggregated and aggregated views and provides zooming functionality to quickly navigate and understand long sequences at large scale. Let me know if you want to try it!

When you’re having a great Friday.

The program of the Causal Data Science Meeting 2023 (Nov 7–8) is now available on our newly redesigned website. Check out the fantastic list of presentations and register for free under causalscience.org 🔊 Please share with your network!

Tons of folks seem to have accounts here now so I don't quite understand why the bsky feed is so sparse. Seems like either nobody I know is posting or something is broken with generating the feed? Likewise my posts here don't seem to get much feedback.

This is one of the most interesting applied research projects I've ever come across! It starts a simple observation: why don't RGB colors on screen mix the same way that paints in the physical world do? Then a lot of interesting, yet accessible, math to explain why and correct the problem 💙💚💛

I never learned to use a fork with my left hand to eat anything even slightly challenging, which is a source of great embarrassment for me at fancy restaurants.

Not sure there’s ever been a joke with enough durability to warrant printing it on a t-shirt.

Packing for a trip is building a classifier for items you’re going to need. False positives are things you lugged the whole time and didn’t use. False negatives are where you have to get creative: having to buy stuff, wear something again, do laundry, or borrow from a friend.

Oh! She created a Ross-Chernoff glyph! www.tableau.com/research/pub...

HR: Here's your on-boarding paperwork! Data Person: What's this extra stipend? HR: That's your mental health stipend! Data Person: Does everyone get that? HR: No, just data folks Data Person: HR: On account of our data quality Data Person: ...is it too late to run? HR: Yes.

Touched some grass last weekend.

I desperately need you all to post more here. I promise I’ll like it all.

(X'X)^-1 X'Y

I wrote about how at Motif Analytics we designed a new domain-specific language for working with event sequence data: we call it the Sequence Operations Language (SOL). We were inspired by many data manipulation languages (particularly dplyr) and tools for working with strings (RegExp) and files.

Guess I gotta start posting here.

Had fun making this episode of What’s New In Data with @seanjtaylor.com from Motif Analytics. We dive into data science, experimentation, funnels (with SQL??) and more https://open.spotify.com/episode/4Bt2oPRI2PlhPXUtHmKMja?si=Nribl563SOmOmbN3tducDA

What is the realistic path for a Bluesky client that has DM capabilities?

Prettiest weekend of the year so far in the Bay Area. I love hanging out after the sun goes down and not freezing my butt off.

$44 billion for Twitter. What a waste. think about how many supreme court justices that money could have bought

being a former academic has broken my brain so badly, if I think of a joke, I always check if someone else has already made it on Twitter before and then if so I bail because it's not a strong enough contribution to the literature.

My best current understanding of the Girl Scouts is it’s an organization designed to skirt child labor laws.

Me: If Twitter dies, I think I might be done with social media Also me: oh cool a bluesky invite