I wonder how many of those posts were ‘this question has been asked already”? And I wonder what stats nerds are doing with all that pent up snark energy?

Demetri@phdemetri.bsky.social · 5d ago

This is the blockbuster of the AI generation

Bauhaus Milhaus

My take on p-values: 1. p-values do what they are designed to do, which is not what most people using them imagine them to do. 2. Trying to fix p-values is a trap, because no single-number summary can do what people want p-values to do! 3. Trying to explain p-values is also a trap, because...

Julia M. Rohrer@dingdingpeng.the100.ci · 2w ago

“A p-value is the most widespread statistical concept that is not definable in a single sentence“ CHECKMATE

Fu*k yes!

Emily Riederer@emilyriederer.bsky.social · 2w ago

Solved a problem the old school way last night (following a Stack Overflow that linked to another Stack Overflow than linked to a GitHub repo with three stars), and I’m going to be riding that high all week long

I’m just going to start imputing missing data and see if anyone notices.

Here's a useful page with a **ton** of common tasks in #RStats, done with parallel syntax for tidyverse vs. data.table vs. base. arelbundock.com/posts/dt_tb_... #rdatatable #tidyverse

By default, brms::loo(..., reloo = TRUE) resets the core count to 1 for sampling the chains within each refit. why you gotta do me like that, #brms? who wants to reset the # of cores to 1? WHO? if you're out there, please let me know why? #rstats

This is amazing and I hippo it gets used widely.

Jamie Cummins@jamiecummins.bsky.social · last mo.

Today we launch the first stable release of RegCheck: v1.0.0. RegCheck makes it easier and quicker to compare study registrations to published papers for consistency - something we know is important in principle, but rarely done in practice. A 🧵 on what's new: regcheck.app

I mean, every stats text too, tbh.

Stephen Nuñez@socio-steve.bsky.social · last mo.

econometrics books should have a big *WARNING: YOUR DATA WILL NOT ARRIVE IN A FORM THAT ALLOWS THIS ANALYSIS. IT MAY TAKE WEEKS TO GET IT THERE * sticker at the top of every section introducing a a new quasi-experimental method.

I feel seen.

Jarrett Byrnes@jebyrnes.bsky.social · last mo.

If you fell off your Mapquest directions in Boston or its environs, you just gave up and prayed you could find a way home.

The internet is awash with hidden landmines—beware. I have stepped on this one and I think I’m on a list.

Stats For Bios@statsforbios.bsky.social · last mo.

Googling INLA (integrated nested Laplace approximation) without care gives you the Irish National Liberation Army. The militia for you if you think the IRA are too centrist.

Holy cow, this is *amazing* and I highly recommend it to anyone who works with environmental exposure data!

Stats For Bios@statsforbios.bsky.social · 2mo ago

Want to estimate the effects of hourly environmental exposures, but only have daily case data (i.e. the normal situation)? Nidhi Shukla developed a method and did a big simulation study to test it. It works pretty well. www.medrxiv.org/content/10.6...

I’m sure I’m not the first to say this, but academic publishers—months to years from writing to acceptance, and the submission related work. And then: return this proof in 2 days (sent on Friday evening, btw). 🤬

Anybody have a resource they like for for fitting frailty models in #brms? Seems like there are some differences in sampling from posteriors. Looking to jumpstart my learning. #rstats

Holy flirking schnitt. Endurance sports nutrition is OUT OF CONTROL. The number of absurd claims and dollars per calorie are inconceivable.

can't argue with this.

Stern Lecture Statistical Consultants. ‘I told you not to do that’

Chelsea Parlett@chelseaparlett.bsky.social · 2mo ago

If me and @phdemetri.bsky.social ever started a joint stats consulting business this would be the tagline

I thought I was the only one!

a cartoon character with blue hair says so this is what it feels like when doves cry ..

ALT: a cartoon character with blue hair says so this is what it feels like when doves cry ..

media.tenor.com

Lana Sinapayen@sinalana.eurosky.social · 2mo ago

I had to write this exact line 2 days ago lol

Every few months i discover that tidyverse has built-in functions to do things I was already doing, just with more steps. I recently discovered value_fill = in pivot_wider. this used to frustrate me, but I’ve come to love it. The breadth of knowledge and expertise out there is 😍. #rstats

I was today years old when I first received this error message: ‘no more error handlers available (recursive errors?). #rstats

It’s both amazing, terrifying, and a little sad that I learned so much regex, got pretty good at it, and now appear to have basically dumped it almost completely from my brain #rstats

I mean, what are we even doing here. pubmed.ncbi.nlm.nih.gov/41661604/

It’s funny cause it’s true.

Marko Bajlovic@marko.tech · 3mo ago

These are my people.

Daniel Lakeland@dlakelan.mastodon.sdf.org.ap.brid.gy · 3mo ago

A lot of people like to play computer games. I like to play computer games too. My favorite is dump a couple gigabytes of CSV files on my computer and let me fuck around until I find out.

Massive deadline approaching, way behind on analysis. Perfect time to update my version of R and RStudio. Duh. #rstats

I’ve had to resort to eating black licorice as it’s the only candy my locust-like children do not devour immediately upon locating in our house.

#brms people. in a random intercept model, prior = c( prior(normal(0, 1), class = "Intercept"), prior(exponential(1), class = "sd")) is equivalent to Aj(mu, tau), mu ~ normal(0,1), tau ~ exponential(1)

Agree with all 3 points. And *hard* agree with #3.

Stephen Wild@stephenjwild.bsky.social · 4mo ago

You may not like this study and its methods may not be sound, but I do not think it is p-hacking. Poor quality != p-hacking, just like high quality != absence of p-hacking Also I do not like the term p-hacking.

The more I read about Bayesian modeling, the more it seems…almost wrong to *not* do it. Especially as samplers, computing power, etc. and especially when one follows a principled workflow as promulgated by @rmcelreath.bsky.social and @avehtari.bsky.social #rstats #bayes

🎯So many important points packed into this editorial. ht @maartenvsmeden.bsky.social www.nature.com/articles/s41...