Kazuki Motohashi

🚨 Due to high demand, I created a new (and much improved) version of the Econ Journal Matchmaker. New tool here: …ed75b671.eastus.azurecontainerapps.io #EconTwitter Please help spread the word by reposting. 🙏 More details below 👇

📊 Based on my journal-submission log since Dec. 2013, I made this figure showing how often manuscripts are rejected—and how long it takes—before (conditional) acceptance. I hope it helps PhD students and junior scholars see the real world of publishing in political science. 💪📚

📢 Just accepted in #JAERE! 📢 "Electricity, Agricultural Productivity, and Deforestation: Evidence from Brazil" by Dimitri Szerman, Juliano Assuncao, Molly Lipscomb, and Ahmed Musfhiq Mobarak. 📎Read it here: buff.ly/wlW7nCy

Another plug for my short practical guide to getting started with Claude Code. I hope it’s useful!

What happens when generative AI meets Stata? Explore community-contributed tools, intelligent bots, and creative tutorials showing how users are integrating LLMs into their workflows. Faster coding. Easier debugging. New possibilities.
 Read more ↓ stata.com/stata-news/new…

🚨 Are you about to submit an Economics paper? I made a free tool that suggests journals/editors that may be a good fit. Super easy to use. Just upload a PDF of your paper: chatgpt.com/g/g-69fbdb3e6a… #EconTwitter 1/

I just released a free 4 hour course on OpenAI Codex. The course focuses on the Codex Desktop app - which is the best interface for agentic coding with the best agentic coding model: gpt-5.5. I assume zero experience with agentic coding, and help you understand fundamentals such as Skills, MCPs, Plugins, Git, Subagents, and Worktrees. However, I've developed the course so that you're doing practical things with Codex within the first 5 minutes. The theory I present is *always* paired with immediate practical payoff. As a final capstone, I show you how to use Codex effectively for the ideation/brainstorming/planning stages of a project, and then I use this process to develop a novel web app from scratch. The full course is available here: youtu.be/j7d5rs0iMlE I've also created a 229 slide deck mapping to the course. You can find this slide deck in the Classroom section of my Skool server "The AI MBA": skool.com/the-ai-mba

Brand-new mini-series: this is an 8-parter on learning models in economics. Topics include: - What is learning - What is Bayesian updating - Generalizations beyond Bayesian updating - Multidimensional learning - Recent empirical papers about learning youtube.com/watch?v=zrQVqq…

I gave a training today on Claude Code to a group of 20 economists. A few takeaways: > Sonnet 4.6 on medium thinking did a pretty good job of converting a paper to Beamer slides for an upcoming NBER talk. I was pleasantly surprised. esp important to be aware if you're on the $20/mo plan > nevertheless, some light usage with few subagents ate through usage limits on the $20/mo plan. IMO Codex with gpt-5.5 has gotten good enough for writing that you should be using it as agentic coding tool number 1 as a default over Claude Code for all econ use cases. I have started using Codex exclusively for writing other than brainstorming/planning docs. > I demoed some "proof" (github.com/EveryInc/proof…) based workflows to iterate on decks/writing between LLM and human. this was very popular - I'm excited to see the evolution of these interfaces. > Claude Code's desktop app is great, and it's what i suggest to most people, but Codex's is a fair bit better thought out and less buggy than Claude Code's. The economists I was teaching quite liked the Claude desktop app - I'll probably make a distinct video on it as a supplement to my Claude Code course (youtu.be/HR1lI4V0oKE?si…) which I made before the desktop app became decent.

IV-PPML-HDFE. We are happy to introduce an instrumental-variable Poisson pseudo-maximum likelihood estimator with high-dimensional fixed effects. We also provide a robust and user-friendly ‘ivppmlhdfe’ package for Stata and Julia. More details here 👇 ideas.repec.org/p/drx/wpaper/2…

By popular demand, here is my Beamer theme that allows annotation of text, figures, and tables using arrows and handwritten-like text: github.com/ambropo/Jambro… A 🤓 thread on what the theme allows to do 👇🏼

I put together a short practical guide for economists who want to use Claude Code, but who haven't gotten around to trying yet. The goal is to reduce the start-up costs by using Claude Code within VS Code.

I am seeing tons of people react to this and say “Look at the AER, QJE” tons of papers with data from developing countries. I did my first 13 years as an academic in Brazil, of course it’s possible to publish good work from the periphery, but it’s SO MUCH HARDER, you have no idea

Excited to share that my paper on sanitation investments in India and their effects on water quality and health is forthcoming in the Journal of Development Economics. 🔗 Paper: doi.org/10.1016/j.jdev…

Want to learn how to visualise and analyze spatial data? My #R #GIS courses are coming up in Apr-May 2026: Intro to using R as a GIS - No prior knowledge needed! Advanced R as a GIS - For those ready to dive deeper nickbearman.com/training-cours…

Just finished recording a ~4 hour course on Claude Code for beginners About half is edited already - should be out on YouTube early next week! Subscribe if you haven't already and you'll be the first to know: @aniketapanjwani" target="_blank" rel="nofollow noopener">youtube.com/@aniketapanjwa…

In the last few weeks I have been working on automating economics research. With the current advances in AI coding agents, many are trying to produce high-quality economics research at scale by using autonomous swarms of AI agents, but as far as I can see nobody did crack the code yet. Therefore, I decided to take it on me to design a sophisticated orchestration system that completely automate economic research from idea to full research article. And I think I have done it. Research in economics will never be the same. The system still needs some tweaking, and it is very expensive, however at this point I can reveal that it produces AER-level research output. I link here the first paper produced with this new system. I would be grateful for your comments. dropbox.com/scl/fi/n82e5q2…

Starting today you can use Codex in Claude Code 👀 /plugin marketplace add openai/codex-plugin-cc Try it out today with: /codex:review for a normal read-only Codex review /codex:adversarial-review for a steerable challenge review /codex:rescue to let codex rescue your code Enjoy Codex-ing!

The great Paul Goldsmith-Pinkham did a series of videos with Markus Brunnermeier on Claude Code (more to come)

I made a skill for PDF reading optimized for academic articles, including tables. As a test case, I ran it on Acemoglu Johnson Robinson (2001). You can see the benchmark results here: github.com/aniketpanjwani… Under the hood - it uses IBM's docling with various optimizations/tweaks for tables (you can see all the tables here): github.com/aniketpanjwani… > thx @EconTraina for docling recommendation You can install it from my skills repository - github.com/aniketpanjwani… - which also includes a Python Learning Skill (x.com/aniketapanjwan…) and has various install instructions Or just point CC/Codex at github.com/aniketpanjwani… and ask it "install this skill" I've got several ideas of how to improve the skill, but feel free to submit a Github issue and/or PRs with improvements.