Channel Title: The Causal Mindset Channel Description: Dr. Quentin Gallea expertise in causal inference has been successfully implemented in research projects addressing timely and complex challenges, such as the effect of COVID lockdowns on virus spread and the impact of weapons imports in Africa on migration flows and conflicts. Moreover, Quentin has extensive experience in teaching statistics and causal inference to diverse audiences, from economists to participants in startup incubators, including medical researchers, engineers, and marketing experts. His practical approach has empowered over 10,000 students to apply causal thinking in their respective fields. Today, Quentin is a Senior Advisor at Enlighten Advisory, offering his expertise to support strategic decisions. <video> The Science and Art of Causality (part 2) Description: There are no test to prove that your effect is causal. Despite this challenge, we can actually do a lot to find evidence, eventually leading to the proof of a causal effect. In this article, I will show you how we could proceed with a fascinating research paper: 'LONDON FOG: A CENTURY OF POLLUTION AND MORTALITY, 1866-1965'^1 Put you in the shoes of a police detective, because that's exactly what we are doing when we want to find a causal effect. We have a suspect. But when we look closer, was it really colonel mustard with the candlestick in the conservatory? Are you sure it was not with a wrench or the crime done by someone else? In addition to the RDD, I wanted to show you also event study. We will use the data from Hanlon (2018): LONDON FOG: A CENTURY OF POLLUTION AND MORTALITY, 1866-1965. London was a densly habited and heavily polluted area already during the 19th century. Fog reports are consistent through the century and allow to capture day where the air was heavily polluted (fog keep the pollution low). Data: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/JD5ZHT Paper: https://www.nber.org/papers/w24488 Subscribe and click the bell! ------------------------------------------------------------ Join the discussion: LinkedIn: https://www.linkedin.com/in/quentin-gallea-phd-74641b56/ Instagram: https://www.instagram.com/stats_with_quentin/ TikTok: https://www.tiktok.com/@stats_with_quentin </video> <video> ChatGPT crash course trailer [2024] Description: Access the full class here today: https://quentin-gallea.systeme.io/4e570836-fe47b441-aa536f8e-f9549395 Harvard Business School showed that consultant using ChatGPT worked faster, produced more output and with a 40% higher quality. Don't you also want to be more productive and have more time to focus on what matters? In this short video, I will address the main myths that stop many people from using ChatGPT (or similar tools) or not use it efficiently. - ChatGPT is Dumb - ChatGPT makes mistakes - ChatGPT's output is generic </video> <video> ChatGPT crash course [2024] Description: Access the full class here today: https://quentin-gallea.systeme.io/4e570836-fe47b441-aa536f8e-f9549395 Harvard Business School showed that consultant using ChatGPT worked faster, produced more output and with a 40% higher quality. Don't you also want to be more productive and have more time to focus on what matters? In this short video, I will address the main myths that stop many people from using ChatGPT (or similar tools) or not use it efficiently. - ChatGPT is Dumb - ChatGPT makes mistakes - ChatGPT's output is generic </video> <video> Exploratory data analysis: My secret tips [in python] Description: After years of doing scientific research, I ended up creating a recipe to start any research project. This recipe, that I am about to share with all of you, helps me every day to have a very structured approach, reducing the risk of missing something and maximizing efficiency. 1:10 The 5 steps 1:58 The example 2:28 1. Variables selection 2:45 2. Sample selection 5:01 3.1 Univariate analysis [Concept] 6:09 3.2 Univariate analysis [Example] 8:23 4. Bivariate analysis 12:29 When the magic happens 12:50 5. Conclusion Link to the article: https://towardsdatascience.com/a-recipe-to-empirically-answer-any-question-quickly-22e48c867dd5 ----------------------------- Join the discussion: Subscribe and click the bell! LinkedIn: https://www.linkedin.com/in/quentin-gallea-74641b56/ Instagram: https://www.instagram.com/stats_with_quentin/ TikTok: https://www.tiktok.com/@stats_with_quentin </video> <video> Can you really question science with my First Aid Statistics? Description: In this video I use two examples to illustrate how the tools I share allow you to challenge even published papers in peer-reviewed scientific journals. Thanks Pr.Joerg Dietz for those two nice examples! Pr. Joerg Dietz invited me to teach during his class Evidence Based Management for the Master in Management at HEC Lausanne. I've heard about those two papers during his introduction of the class. 2:00 1st example: Is there something else? (omitted variable bias) 4:38 2nd example: Can we extrapolate? (external validity) 7:00 in-vivo research example Article 1: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0236131 Article 2: https://www.sciencedirect.com/science/article/pii/S0959475217305716 ----------------------------- Join the discussion: Subscribe and click the bell! LinkedIn: https://www.linkedin.com/in/quentin-gallea-74641b56/ Facebook: https://www.facebook.com/Unbiased-making-numbers-speak-111123366941359 </video> <video> Netflix stock crash: What can we learn with statistics? Description: Netflix is not chilling. 24% drop of the stock in few hours. Let's see the 2 things we can learn from this. What's the CAUSE? (of course I talking about causality, not finance :) ) Many people around the internet argue that it's caused by what the same people would call "wokeness". BUT, Could it be something else? (first tool I present from my first aid statistics in my TEDx: https://youtu.be/bD1Jq6YwPk8). Could it be rather the increased competition by other companies, or the fact that they exploring a version with ads? We have just one Netflix and we lack the counterfactual, the world with netflix but without increased competition or without version with ads, or with different content. So it seems impossible to claim what is (are) the cause. However here is what we can learn two things and also reflect what we should do when something similar happens to our company: 1:10 1st Lesson: Sudden drop, sudden cause 2:42 Causality is everything 3:42 Real underlying cause? 4:20 2nd Lesson: everything is multifactorial 5:35 What can we do (when we face such a situation with our company with our projects)?! ----------------------------- Join the discussion: Subscribe and click the bell! LinkedIn: https://www.linkedin.com/in/quentin-gallea-74641b56/ Facebook: https://www.facebook.com/Unbiased-making-numbers-speak-111123366941359 </video> <video> Why causal inference? Description: </video> <video> The Science and Art of Causality [part 1] Description: Let's go together on a journey towards answering one of the most valuable question you should ask yourself: How to assess causality? Every choice you make or decision taken is usually the results of suspected causal relationships. For example: Individual choices: If I go vegan, I'll reduce my ecological footprint. If I drink this tequila shot, I'll dance better. Companies: Home-office reduces productivity Spamming users with YouTube Premium Ads will increase the number of subscribers. % fines if parents take children late from nursery Policy makers: Replacing nuclear power plants with renewables will help to reach the Paris Agreement. Lockdowns will reduce the spread of the covid-19 However, there are no test to prove that your effect is causal. Despite this challenge, we can actually do a lot to find evidence, eventually leading to the proof of a causal effect. Let's see together what can we do. Subscribe and click the bell! ------------------------------------------------------------ Join the discussion: LinkedIn: https://www.linkedin.com/in/quentin-gallea-phd-74641b56/ Instagram: https://www.instagram.com/stats_with_quentin/ TikTok: https://www.tiktok.com/@stats_with_quentin </video> <video> Understanding the average in statistics: from variability to Simpson's paradox Description: The average is arguably the most common measure used in statistics. If you think that it’s a simple concept, think again. It’s simple to compute, but also very simple to use to hide information, to mislead. Let’s dive together to see what might be hidden behind this common measure. It the worst case scenario, the mean can represent the opposite of the truth (in some sense). 1:03 Behind the mean 2:54 Simpson's paradox 5:37 Behind the mean ----------------------------- Join the discussion: Subscribe and click the bell! LinkedIn: https://www.linkedin.com/in/quentin-gallea-74641b56/ Instagram: https://www.instagram.com/stats_with_quentin/ TikTok: https://www.tiktok.com/@stats_with_quentin </video> <video> Believe It or Not: How Sora AI Changes the Game for News, History, and Hoaxes Description: Imagine a world were you can't rely anymore on visual proofs, where videos shown on the news have no value anymore. This scenario isn't just hypothetical; it might be our near future 🚀 OpenAI just released SoraAI a text to video AI able to create perfect video shots based on prompt. Distinguishing real from virtual images might be soon virtually impossible. Imagine the news of tomorrow. What will be the consequences for us, for the future generation, for fake news, for our perception of reality? 👾 Remember the bigfoot, UFO or lock ness monster images? All blurry and in terrible quality which was a major argument why people would not believe it. 💥 Imagine now perfect war shots, speeches, riots, etc. You could have different sources with different quality visual "proofs". How can you deal with this? What is real? This non trivial question is becoming more and more complex to answer with the exponential growth of AI tools. </video> <video> ChatGPT crash course trailer [2024] #chatgpt #ai #openai Description: Access the full class here today: https://quentin-gallea.systeme.io/4e570836-fe47b441-aa536f8e-f9549395 Harvard Business School showed that consultant using ChatGPT worked faster, produced more output and with a 40% higher quality. Don't you also want to be more productive and have more time to focus on what matters? In this short video, I will address the main myths that stop many people from using ChatGPT (or similar tools) or not use it efficiently. - ChatGPT is Dumb - ChatGPT makes mistakes - ChatGPT's output is generic </video> <video> Where to find the new content and future of the channel Description: Hey legends, I'm diversifying my productions. I'll focus on micro content, weekly memes and post, with very short videos (less than a minute), on Instagram @stats_with_quentin ( https://www.instagram.com/stats_with_quentin/ ), while here I'll produce macro content, longer video (7-15min) with potentially more advanced content. Let's grow this thing together! I need you! Cheers! </video> <video> Causal Inference for Everyday - Question Claims from OpenAI to Trump Description: Welcome to the first edition of the "Weekly Causal Insights", where we question causal claims in the most significant news from last week. To do so, we apply the framework, The Causal Mindset”, which is a set of accessible and practical tools grounded in causal inference (the scientific way of questioning causality). My goal is to empower you against abusive claims, misinformation and manipulation, and hence support your decision making. This newsletter not only helps you to separate facts from fiction from past events, but also teaches you how to spot future issues. As a bonus, this newsletter offers rich real-life examples for people teaching causal inference. This first issue contains: 1. Politics & Media Influence: "From Elon Musk to Carlos Espina: Meet the Influencers Shaping the US Election" 2. Sports & Social Dynamics: "Raygun and Australian Officials Speak Out Against ‘Hate’ Over Olympic Breaking" 3. Geopolitics & Military Strategy: "Ukraine's Invasion of Russia is Showing How Good it is at Hitting Putin's Weak Spots" 4. Technology & Power Dynamics: "Mark Cuban Says 'Everybody's Chasing Power, And Nothing Will Give You More Power Than Military And AI'" 5. Cybersecurity & AI Ethics: "OpenAI Says It Caught a ChatGPT-Powered ‘Iranian Influence Operation’" If you want to apply these concepts to your own situations or articles, head to the https://causalmindset.com to access the free companion App. ------------------------------------------------------------------------------------------------------------- 🤝 Connect with me on: [LinkedIn] [Instagram] [Medium] https://www.linkedin.com/in/quentin-gallea-phd/ https://medium.com/@quentin.gallea https://www.instagram.com/the_causal_mindset/ ------------------------------------------------------------------------------------------------------------- About Quentin Gallea: Dr. Quentin Gallea expertise in causal inference has been successfully implemented in research projects addressing timely and complex challenges, such as the effect of COVID lockdowns on virus spread and the impact of weapons imports in Africa on migration flows and conflicts. Moreover, Quentin has extensive experience in teaching statistics and causal inference to diverse audiences, from economists to participants in startup incubators, including medical researchers, engineers, and marketing experts. His practical approach has empowered over 10,000 students to apply causal thinking in their respective fields. Today, Quentin is a Senior Advisor at Enlighten Advisory, offering his expertise to support strategic decisions. </video> <video> How to learn causal inference on your own for free [2024] Description: Here it is finaly, the answer to the question I've been asked the most about online: How to learn causal inference? Where should I start? What are the concepts I should understand and in which order should I learn them? Mastering causal inference will give you a tremendous asset on the job market and beyond. And good news, you'll find below my ultimate self-study guide for all level and necessitate no pre-requisite. I crafted carfully this guide over the last decade, while doing scientific research, consulting, and teaching causal inference to over 10,000 students. Written guide with all the links available here: https://medium.com/towards-data-science/how-to-learn-causal-inference-on-your-own-for-free-98503abc0a06 ------------------ Timestamp and Resources: 1:17 Overview 2:59 1. Introduction: Causal inference key concepts 4:08 1.1 The fundamental problem of causal inference 📚Resources: 🌐TEDx: How to question numbers and prevent manipulation (15min) https://www.youtube.com/watch?v=bD1Jq6YwPk8&ab_channel=TEDxTalks 🌐 The Science and Art of Causality (part 1) https://towardsdatascience.com/the-science-and-art-of-causality-part-1-5d6fb55b7a7c 🌐 The Causal Mindset App https://www.thecausalmindset.com/ 5:10 1.2 A little bit of formalization: Potential Outcomes 📚Resources: 🌐 Short intro: The potential outcomes (Causal Inference for the brave and true) https://matheusfacure.github.io/python-causality-handbook/01-Introduction-To-Causality.html 🌐 In-depth: The potential outcomes (Causal Inference: The Mixtape) https://mixtape.scunning.com/04-potential_outcomes 5:30 1.3 Visual representation of causal links: Directed Acyclic Graphs 📚Resources: 🌐 Short intro: Directed Acyclic Graphs (Causal Inference for the brave and true) https://matheusfacure.github.io/python-causality-handbook/04-Graphical-Causal-Models.html 🌐 In-depth: Directed Acyclic Graphs (Causal Inference: The Mixtape) https://mixtape.scunning.com/03-directed_acyclical_graphs 5:53 2. Technical Tools 6:03 2.1 Probability, Statistics, and Linear regression 📚Resources: 🌐 Probability and Regression Review (Causal Inference: The Mixtape) https://mixtape.scunning.com/02-probability_and_regression 🌐 In-depth: Directed Acyclic Graphs (Causal Inference: The Mixtape) https://matheusfacure.github.io/python-causality-handbook/05-The-Unreasonable-Effectiveness-of-Linear-Regression.html 🌐Which variable you should control for? (Scott Cuningham’s substack) https://causalinf.substack.com/p/which-variables-do-i-need-to-control 🌐A Crash Course in Good and Bad Controls (Cinelli et al. (2022)) https://ftp.cs.ucla.edu/pub/stat_ser/r493.pdf 🌐Fixed Effects (Causal Inference for the brave and true) https://matheusfacure.github.io/python-causality-handbook/14-Panel-Data-and-Fixed-Effects.html 🌐Fixed Effects (Causal Inference: The Mixtape) https://mixtape.scunning.com/08-panel_data 7:00 2.2 Statistical tools 📚Resources: 🌐 Introduction To Python Programming (Udemy) https://www.udemy.com/course/pythonforbeginnersintro/ 🌐 Linear Regression in Python (Real Python) https://realpython.com/linear-regression-in-python/ 8:02 3. Randomized Experiments (A/B testing) 🌐 Randomised Experiments (Causal Inference for the brave and true) https://matheusfacure.github.io/python-causality-handbook/02-Randomised-Experiments.html 9:53 4. Quasi-experimental design 10:04 4.1 Regression Discontinuity Design (RDD) 📚Resources: 🌐Short intro: Regression Discontinuity Design (Causal Inference for the brave and true) https://matheusfacure.github.io/python-causality-handbook/16-Regression-Discontinuity-Design.html 🌐In depth: Regression Discontinuity Design (Causal Inference: The Mixtape) https://mixtape.scunning.com/06-regression_discontinuity 12:28 4.2 Difference-in-Differences (Diff-in-Diff) 📚Resources: 🌐Short intro: Diff-in-Diff (Causal Inference for the brave and true) https://matheusfacure.github.io/python-causality-handbook/13-Difference-in-Differences.html 🌐In depth: Diff-in-Diff (Causal Inference: The Mixtape) https://mixtape.scunning.com/09-difference_in_differences 🌐Example: ChatGPT effect on Stack Overflow https://towardsdatascience.com/how-ai-is-changing-the-way-we-code-36ff30262e65 12:56 4.3 Instrumental Variables 📚Resources: 🌐Short intro: Instrumental Variables (Causal Inference for the brave and true) 🌐In depth: Instrumental Variables (Causal Inference: The Mixtape) 12:23 5 Your first meaningful project in causal inference 16:57 6. Advanced topics Check my Towards Data Science articles for more resources ------------------------------ 🤝 Connect with me to learn more You can find me on Linkedin (main platform where I post every week), Instagram, YouTube, and of course, you can follow me here on Medium. If you want to have me for a talk, teach a class, a seminar or give a training, you can contact me: quentingallea.com </video> <video> How I'd Learn Causal Inference (If I Had to Start Over) Description: Here is my full self-study guide using exclusively free online resources for causal inference. Head out to my channel https://www.youtube.com/watch?v=XBJI-pQKDDY </video>

Contacts and Links

Description

Dr. Quentin Gallea expertise in causal inference has been successfully implemented in research projects addressing timely and complex challenges, such as the effect of COVID lockdowns on virus spread and the impact of weapons imports in Africa on migration flows and conflicts. Moreover, Quentin has extensive experience in teaching statistics and causal inference to diverse audiences, from economists to participants in startup incubators, including medical researchers, engineers, and marketing experts. His practical approach has empowered over 10,000 students to apply causal thinking in their respective fields. Today, Quentin is a Senior Advisor at Enlighten Advisory, offering his expertise to support strategic decisions.

Videos from channel