Noooo don’t overfit your so sexy hahah

I love nothing more than when my coworkers’ dogs walk in frame during a zoom meeting 🥹

“look how much better my favorite model is compared compared to these other TRASH MODELS😒😒😒 when I specifically chose a DGP that matches my models assumptions and not the others’😌”

LET HIM COOK

If no one values you spending time on documentation, just tell them you’re making a Chatbot and you’re building content for RAG

🪧

ordbetareg has now fallen out of the top 10% of CRAN downloads. I mostly blame @chelseaparlett.bsky.social for not teaching enough classes to enough students in order to properly inflate my download statistics.

BLUSKY IS THIS BOOKMARK I made for Tony’s urbanism group too morbid?? We have a booth at a dinosaur themed children/family event👀

I fell off my bike on Saturday and got a gash on my forehead (I’m okay, helmets are good!!) I could handle the pain, but now I’m in the itchy stage and I’m going INSANE

Writing about statistics is exhausting because we’re (statisticians) all so persnickety about everything

sometimes, when my engagement is low, i think to myself "should I start shit posting about p-values again for attention?"

Frisch WAH Lovell theorem

If you’re struggling to understand CUPED for A/B testing, but you already understand why a paired T-test has higher power than a two-sample T-test on the same data, congratulations you actually do understand the important parts of CUPED!

Does JavaScript….go hard?!??

Explaining things to people who are learning about them for the first time is the BEST WAY to strengthen your own knowledge and communication skills🤯 You 1) figure out what you don’t know well enough to explain 2) get incredible questions that help you understand what’s not intuitive to newbies

Me thinking about survival analysis: 🎶my estimator has a second name it’s M-E-I-E-R🎵

🤨 STOP IGNORING UNCERTAINTY IN YOUR FORECASTS In this blog I wrote about how Continuous Ranked Probability Scores (CRPS) capture two important ways that a probabilistic forecast can be good: ☝️its correct ✌️its precise getrecast.com/crps-evaluat...

Made these super cute bookmarks to give out at Tony’s urbanism group event tomorrow 🥹

Me trying to figure out why LLMs keep giving me incorrect answers to my stats questions 🫩 (OG 🎃 pic from @qian.dreaming on insta)

This is one of my biggest fear with LLMs. They reduce the incentive to create new, non-LLM materials, so we’re going to train on more and more LLM material and reduce the diversity of style/content/topics in the world (Pic from Wikipedia)

Hey CA, you got your ballot in? Times running out. A zillion boxes out there - one's on your path today.

It’s such a great feeling to re-visit an article/blog/book that *used* to feel confusing and overwhelming and feel like it makes total sense now😎

Hell yeah. More politicians like this.

The worst part of working in marketing measurement is that text editors never recognize “incrementality” as a word

We’ve discussed who owns the *means* of production, but what about the *skews* and *kurtosis* of production? 🤔

when i think about all the important moments in my life, and try to guess which will flash before my eyes at the end, my best guess is that it will be kurtosis🤔

Surprise I have been working at Canva for over a month, tackling some incredibly interesting and challenging problems in Marketing Analytics and Measurement 😍 Very excited to be a Staff Data Scientist “Making Complex Things Simple” on such a collaborative and brilliant team🥳

Felt like a great day to check out Braiding Sweetgrass from my library🥳 I loved the Serviceberry by Robin Wall Kimmerer, so I can’t wait to listen to this🌿

Sometimes I think about the fact that some of my jokes are things people actually say un-ironically on LinkedIn 🙃

My job would be so much easier if correlation was causation😔

Measurement?? You mean COUNTING things?? 🤨

Thank you Gavy 🥳 And a bigger thank you to all the people from Senator Wiener, to @streetsforall.org, to all the people who have been yelling about this from (our currently less-dense) rooftops 🤩

Me trying to convince junior data scientists that you have to master the basics first so that the more complex stuff is easier down the line

NYC based instructors: we're looking for an adjunct for a graduate level Ethics of Data Science class this Spring. Please share widely apply.interfolio.com/175474

I've been thinking about developing this course for a little while, and the @chelseaparlett.bsky.social post and reaction may be the tipping point for me to finally take it on: bsky.app/profile/chel...

Nova the urbanism corgi is the headline image for the OpEd @tonypelleriti.bsky.social wrote about how newspapers cover public transit vs. car-centric projects 👀 voiceofoc.org/2025/10/pell...

So many things, wonderful collaborations, podcast appearances, consulting opportunities and jobs are indirectly or directly due in part to my social media presence!

Counter factual??? You mean FAKE NEWS???

WE have GOT TO START differentiating between things that are Bayesian “culturally” and things that are Bayesian in principle. I keep seeing people claim “only Bayesian models XYZ” but it’s not inherently true, it’s just easy to implement in bayesian software and harder in frequentist software

There are so many people spouting incorrect information about Marketing Measurement on LinkedIn 🫠 sometimes they’re even fighting with each other but both are incorrect. “Mommmm people are wrong on the internet again”🥴

You can get soooo much attention/delight on LinkedIn just because people don’t typically recognize large powers of 2 (and are shocked and awed when you point out that a seemingly random number in base 10 is a round number in base 2) 🤯 Them:

MC Stan is here! Follow for the latest Stan news, and tag if you want us to repost your posts about new papers, packages, courses, etc. about Stan

It’s not the method that makes you causal it’s the assumptions

When the query is taking longer than I thought, and I’m starting to fear that I might’ve done something wrong

One of the most powerful skills you get with more and more experience is just understanding what you can ignore lol. It’s made reading stats papers much easier 😂

“it’s October 3rd” Happy K-Mean Girls Day to those who celebrate

👋 After a wonderful experience working on one of the best Marketing Mix Models in the biz with some of the most thoughtful and smartest data people I’ve met, my time with Recast came to an end in August👩🏻‍💻 👀 but exciting career update to come 🥳

Be glad I’m not queen of the world because if I was it would be illegal to report point estimates without uncertainty🤺🤺🤺

The hardest part about being a data scientist isn’t the math, it’s explaining results/risks/assumptions clearly to people who aren’t familiar with the math