Carlos E Lourenco (Caê)

New paper online!🎉
We introduce a new concept: AI-related conspiracy theory, which claims that powerful humans use AI for malicious purposes or AI itself eventually controls humans.
Yes, this is inspired by the movie “the Matrix”.👻
bpspsychub.onlinelibrary.wiley.com/doi/full/10....

Analysis of Sales Shift in Retail with Causal Impact: A Case Study at Carrefour | Towards Data Science Applying causal inference to measure the effect of product unavailability on retail sales at Carrefour

ICYMI: Thanh Liêm Nguyen's deep dive into applying causal inference to measure the effect of product unavailability on retail sales at Carrefour.

Not sure it’s quite what you want but making the point that big data doesn’t guard against bias

declaredesign.org/blog/posts/w...

What Good Is a Regression? Inference to the Best Explanation and the Practice of Political Science Research | The Journal of Politics: Vol 87, No 4 We argue that almost all empirical social science research should employ a mode of argumentation called “Inference to the Best Explanation” (IBE). While elements of IBE appear widely, it is seldom con...

www.journals.uchicago.edu/doi/abs/10.1...

Is theory getting lost in the "identification revolution"? The following is a guest post from Columbia University political scientist John Huber, and is a slightly modified version of […]

Not your focus, but you might want to pair it, at least as recommended reading, with the recent Stewart/Sperling piece on IBE IN JOP. I for one find John D. Huber's points here very important, too: goodauthority.org/news/is-theo...

Wow. Fantastic. This seems exactly like what I am looking for thanks!

For others:
www.journals.uchicago.edu/doi/full/10....

Colleagues: What’s your favorite, accessible resource (reading, video, slide deck) to convince students that regression on observational data—even with many controls—is not often causal. (Quasi-)experimental designs are preferred for causal inference.

🚨New Preprint: We develop a novel task that probes counterfactual thinking without using counterfactual language, and that teases apart genuine counterfactual thinking from related forms of thinking. Using this task, we find that the ability for counterfactual thinking emerges around 5 years of age.

The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2025 The Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel 2025 was awarded "for having explained innovation-driven economic growth" with one half to Joel Mokyr "for having identified ...

Congratulations to Joel Mokyr, Philippe Aghion, and Peter Howitt for their well-deserved Nobel Prize in economics!

Their groundbreaking work opened the black box of innovation in relation to economic growth, both being core features of modern economies and human wellbeing

#EconSky

Economics Nobel prize won by researchers who showed how science boosts growth Joel Mokyr, Philippe Aghion and Pewter Howitt share economics prize for work that underlines the importance of investing in research and development.

Joel Mokyr, Philippe Aghion and Pewter Howitt share economics prize for work that underlines the importance of investing in research and development.

go.nature.com/4hj9NNV

And the investigation can ultimately lead to a better graphic which makes the causality clearer.

Note that if you want this, don't say "draw me a chart of..." that probably gets you something imagined from its prior. Here I ask it to tell me a story of the key events that led to the decline of horses in the US and then map those events to real data with a chart in matplotlib.

I am pleased to learn that ChatGPT can now make a chart based on real data and discuss the chart with me. And I can ask it questions like "Does this curve look like it represents what I would naively think or is the underlying causality different?" and it can investigate that and comment on it.

Some personal notes on causal inference.

thestippe.github.io/statistics/c...

Disclaimer: I am a big fan of Rubin's way of thinking, I think it's simple yet very elegant.

PS: if you have any recommended reading on the topic, pls write me here or DM me if you prefer.

Pleased to see our new paper finally out!

Together with joël bühler and Roberto Iacono, we study income mobility dynamics in Norway using 26 years of high-quality register data.

Main result: upward mobility is primarily driven by labor income — and almost never by capital alone. (1/4)

In my new WD paper, I analyze the global capital and labor income distributions, highlighting a capitalization process that benefited the global middle class at the start of the 21st century.

📄 Open-access paper: doi.org/10.1016/j.wo... 

📊 Database & description: www.mranaldi.com/research

surveydown: An open-source, markdown-based platform for programmable and reproducible surveys This paper introduces the surveydown survey platform. With surveydown, researchers can create surveys that are programmable and reproducible using markdown and R code, leveraging the Quarto publicatio...

Excited to announce that our paper with @pingfanhu.bsky.social and Bogdan Bunea on our #rstats package {surveydown} is now published in @plosone.org

The paper compares the benefits of using a code-based approach to survey design, leveraging #quarto and #shiny

journals.plos.org/plosone/arti...

Twenty-Five Things You Didn't Know You Could Do with R Explore 25 novel ways to utilize R, including pulling data from various sources, data visualization, mapping, using Quarto for reporting, automating tasks,.

If you missed the Cascadia R Conference 2025 in June, you can still watch David Keyes' @davidkeyes.bsky.social keynote, "25 Things You Didn't Know You Could Do with R."

It's got info on working with PDFs, Qualtrics, automation, prompt engineering, and more!
#R

rfortherestofus.com/2025/06/twen...

Fun viz by Wondimu Ahmed www.linkedin.com/in/wondimuah... from this week's #tidytuesday data. #r

Analysis of covariance and standardization as instances of prediction - PubMed In this paper, prediction provides the basis for unifying the procedures of covariances adjustment and standardization. Analysis of covariance is a method of forming predictions from a linear model; i...

This looks interesting. I shall check it out.
This pubmed.ncbi.nlm.nih.gov/7171691/
Is a favourite paper of mine.
Turning regression into predictions has been a feature of Genstat ever since the 1980s. Peter Lane did the programming.

✨ Unleash your inner stats sparkle ✨ with this very non-technical introduction to marginal effects Update August 2025: There's now a manuscript by Vincent Arel-Bundock and me providing a more thorough introduction to marginaleffects, Models as prediction machines: How to convert confusing coefficie...

This is in a certain sense the follow-up to a blog post I wrote 3 years ago (!) after I had looked into the topic for a paper with Felix Elwert, who is thus once again to blame for large chunks of my output.

www.the100.ci/2022/05/27/%...

📌

Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities Abstract Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.

A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals). Illustrated are 1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals 2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and 3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.

Ever stared at a table of regression coefficients & wondered what you're doing with your life?

Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...

Interested in models used to estimate lagged effects in panel data? We (@rebiweidmann.bsky.social, Hyewon Yang) have a new paper looking at patterns of stability and their implications for bias and model choice: osf.io/preprints/ps... [1/x]

Another chapter in the book of “Sometimes There’s Nothing There & Using Fancy Stats to Find Something Can Be a Bad Idea”

Do you wish we could understand causality by finding lagged effects in longitudinal data?
Bad news - lagged effects are probably rare, & even the RI-CLPM is an ill-suited test.

Feed: "Statistical Modeling, Causal Inference, and Social Science"
By: Andrew on Wednesday, September 17, 2025

Analysis of Sales Shift in Retail with Causal Impact: A Case Study at Carrefour

Applying causal inference to measure the effect of product unavailability on retail sales at Carrefour

#ai #news

📌

📌

Let’s say you have very low sample response rate (recruit tens of millions and only tens of thousands self select into sample).

Then you have a RCT where there is clean random assignment of treatment.

Does the first stage sample selection bias your estimate of causal effect of treatment?