In the last year, I've been saving and categorizing around 20 research papers and analyses per month in my @NotionHQ
This newsletter will make me look deeper, recall, order and connect those ideas periodically to help me shape better opinions and understand this complex world!
The Software Complexity of Nations
Nice economic complexity analysis of open source data. Suggests that software complexity represents a distinct type of competitive advantage and that countries specialise on programming languages related to those they are already capable on.
Delighted to release this paper on Auten & Splinter’s work, painstakingly going through their code & assumptions
https://t.co/xmhGWDwOcv
We uncover key mistakes—and show that once fixed, their methodology implies the same rise of inequality as in Piketty-Saez-Zucman
🧵
When are the most goals scored in a football match?
I’ve been asking myself this question my whole life, and I finally found the dataset to answer it with @graphext , using 44K goals from International football results from 1872 to 2024. Clearly, the chances increase significantly as halftime and the end of the game approach.
There's a great indicator for predicting how well a country's economy will develop. It is based on the diversity and originality of the products a country knows how to manufacture and export, which @cesifoti and @ricardo_hausman refer to as the Economic Complexity Index.
For instance, Spain manufactures and exports many different and diverse products, but those requiring more specific knowledge (like car engines) are becoming less original as more countries are learning how to manufacture them, leading to potentially lower margins.
On the other hand, South Korea or Taiwan, despite manufacturing less diverse products, have in recent decades learned to produce new technological products like liquid crystal panels that few countries are currently exporting. Therefore, their economies appear to be more prosperous in the future.
At @graphext, we are making the impossible possible: working with datasets of millions of rows in a browser.
We are looking for programmers who know C++ and/or TypeScript to continue building on top of this impossible front 👇 https://t.co/pMXWa9bivW
Accessing U.S. data for research just got easier
New online portal streamlines requests for massive data sets at 16 federal agencies
https://t.co/C7D1j4dcIi
Minorities are stopped more (~30%) and pay more in fines (~30%). Beyond the estimates, our approach can be used to answer related questions and highlights the potential value of private sector data to help inform major social challenges.
#econtwitter
https://t.co/nZgM76TguS
Chart-topping original movies have gone extinct. People have a lot of explanations for this, but they're all incomplete because they don't realize the same thing is happening everywhere. An oligopoly has conquered all of popular culture.
I made this personalized doormat for my home using this great strip from Wiley Miller to never forget we live in a complex world with complex answers and our only hope is keep reading and avoid strong opinions
Super interesting paper on Nature using data from Reddit communities they found what communities are more polarized in socio-demographic terms. By @ashton1anderson and @isaacwaller
What communities are you more surprised by?
https://t.co/EbBj76jHqU
The US redistributes a greater share of national income to low-income groups than any European country... but access to education and healthcare or labor market regulations might explain why Europe is less unequal.
Teach a man to fish!
Read the paper 👉 https://t.co/eL7CaHwrsz
Are parents spending less time with their kids? Dads and moms are spending more and more time than ever before with their children.
https://t.co/LqwbJzcL1s
Thanks to a law passed in California in the 70s where all agricultural pesticide used has to be reported into a public database, epidemiologists have found that there is a clear association with the prevalence of Parkinson's in these areas! https://t.co/ADiCKsgJFl
A simple machine learning model considering the combination of nationality, age, sex and few other factors was 2-4x times more effective detecting asymptomatic infected travelers at the Greece border than the random sampling most nations have been applying for Covid testing 🦠📊
Just published @Nature
May represent the most important successful application of #AI in the pandemic (only a few are on the list) https://t.co/rHchKM606p Reinforcement learning for efficient testing at the Greece border
The Spanish Tax Agency + Social Security + Bank of Spain are joining forces to join their data table and make them available to researchers to make more data driven policy making in Spain 😍📊
El #INE, la Agencia Tributaria, la Seguridad Social y el Banco de España acuerdan comenzar a trabajar conjuntamente en el desarrollo de un sistema de acceso a sus bases de datos con fines científicos de interés público
https://t.co/OeXqNksZq5