Felix

A privacy-first, open-source knowledge base A privacy-first, open-source platform for knowledge management and collaboration.

I use logseq.com. It resembles a daily bullet journal, is open source, has a zotero integration, and to-do list functionality and syncing across devices is possible using syncthing! @obsidian.md is another alternative in that space. :)

Unti AI can give me then article providing a comparison of correlation coefficients across fields, I remain unconvinced. I think it was posted here or on the bad place, but I forgot to save it. Does anybody know the article?

EPFL, ETH Zurich & CSCS just released Apertus, Switzerland’s first fully open-source large language model.
Trained on 15T tokens in 1,000+ languages, it’s built for transparency, responsibility & the public good.

Read more: actu.epfl.ch/news/apertus...

Details on how to access anonymized data from the Creative Destruction Lab, including a description of dataset construction and included variables, from Amir Sariri, Evgenia Gatov, Geneva Neal, Kyle Robinson, Sonia Sennik, Wei Yang Tham, Michael Vertolli, and Av... https://www.nber.org/papers/w34127

I can recommend Espanso for shortcuts. Open-source, easy to set up, and works almost anywhere, so one can insert pipes in any program by simply typing the self-assigned trigger.
espanso.org

I'll throw in "The Effect" by @nickchk.com! Book is available at www.theeffectbook.net and covers a lot by emphasizing the underlying intuition! :)

A laptop in PowerPoint presentation mode on a train.

The "I'm presenting" AOM 2025 badge.

Excited to be heading up to Copenhagen for #AOM2025. One more train, a little sleep and 8-ish more hours until the fun begins.

The picture shows a screenshot of the Stata windows with the QOG dataset loaded. The dataset contains a notes with the quote "Even in a stupid, stupid universe where we have hotdogs for hands, we get very good with our feet." from the movie Everything Everywhere All At Once.

Love to discover these little gems in research 💎💎💎: Someone at the @qoginstitute.bsky.social decided to include this quote from the movie Everything Everywhere All At Once in the notes of one of their datasets.

The sigil was drawn in salt and ash, the candles lit at the pentagram points, the incantation declaimed.

There was a shimmer - a demon appeared.

"Curious. What ritual is this?"

"I got it from ChatGPT. I included all protections in my prompt!"

"I see," the demon said and stepped out of the sigil.

I, for one, am tired of [gestures broadly]. I have grown weary of [points in all directions simultaneously].

Africa's top survey network, @afrobarometer.bsky.social has joined Bluesky:) 🌟🌟🌟Help spread the word.
@dadakim.bsky.social @joeasunka.bsky.social @malawi24.bsky.social

American Journal of Political Science | MPSA Journal | Wiley Online Library Many researchers use unit fixed effects regression models as their default methods for causal inference with longitudinal data. We show that the ability of these models to adjust for unobserved time-...

. @igelstorm.bsky.social, I attended your talk at the 2025 Turing CIIG Seminar and loved it. I wanted to share this article by Imai et al. (2019), which provides estimators when data on exposure is archival and observational. I hope it is of relevance for you! :) doi.org/10.1111/ajps.12417

Every year around now, I start to see grad students on social media bemoan that they’re going home for the holidays to a family that doesn’t understand their research, or what they do in general, and it breaks my heart. 🧵 🧪 #SciComm

I saw a post here a while ago explaining how the author avoided using specific technical terms with non-academic audiences when asked about their work. The point was to illustrate that doing science often consists of learning the right terms. I am looking for that post it was much more eloquent!

We've had a paper reviewed by ERROR, which has concluded we had a "major error that affects a core conclusion"
Few comments on the conclusion and the process in general
1/n

Upgrade your #causalinference arsenal.

A revision of our book "Causal Inference: What If" is available at miguelhernan.org/whatifbook

Thanks to everyone who suggested improvements, reported typos, and proposed new citations and material.

Enjoy the #WhatIfBook plus code and data. Also, it's free.

Thank you! I loved to hear everyone's voices! And @monteiroresearch.bsky.social, you are indeed highly contagious, even for someone who's more inclined towards quant methods! :)

The link is broken for me! :)

Thank you for the pointers @pausalz.bsky.social, @travisgerke.bsky.social, @malcolmbarrett.malco.io! I'm looking for an overview to describe exposure intensity, duration and trends for recurrent events over a certain time period. So I'll see where the pointers will take me!

Maybe @travisgerke.bsky.social or @pausalz.bsky.social can help? Based on my search, I am thinking epidemiology covers via "exposure measurement", but I am missing the right keywords.

I’m looking for a concise introduction or overview of exposure measurement methods for recurrent event data, particularly when the recurrent event data is used as an independent variable. Any recommendations on foundational texts, or key papers are greatly appreciated! #causalsky #econsky #episky

Our Approach:
We first identify where students need support using standardised tests. From there, we offer tailored courses to help them overcome these hurdles.

Our goal is to help students reach their full cognitive potential and thrive in their studies. 3/6

The Challenge:
Many students don’t struggle due to lack of skills but because adapting to university-style learning is tough. It’s about managing time, study strategies, and building focus in a more independent setting. 2/6

🌟Giving this a go: I want to share a bit about our teaching project, "We Help., You Grow" #WHYG from time to time here. The goal is to support students as they navigate the transition to university life—without lowering expectations or standards. #TeachEcon 1/6

Oui!

Have you ever thought to yourself… how can I make my #rstats plot UGLIER? Well, look no further, cuz it’s good plot-bad plot time! My students made plots that will make you want to laugh, cry, and throw your computer out the window. Which one do you think is the worst? 🧵🧪

Sorry for the delay! I just tried it out. It works perfectly. I get the warning message. :) There are more countries with changing ISO codes. I can compile a list, if you would like.

Hi Leo, sure! In 2011, South Sudan declared independence from Sudan. As a result Sudan ISO3 numeric code changed from 736 to 729. If I run isocodes on my dataset and then try to merge it with another dataset that already contains old isocodes, it will not merge the entries for Sudan. Clear? :)

This looks great! I am also working in R, where the excellent countrycode package exists. When working with the numeric ISO code, I encountered changing ISO codes resulting in incomplete merges. For example, Sudan changed from 736 to 729 after the split. An information message would be useful!

Could fastDummies work for your use case? :)