Profile avatar
prashantgarg.bsky.social
Econ PhD @imperial. Visiting researcher at IFC and Cambridge. AI and networks in economics. www.prashantgarg.org
262 posts 5,168 followers 3,662 following
Regular Contributor
Active Commenter
Posts 18 Comments 32

It is so super important to me to see that Health is the only field in economics with substantial rise in null result publications in the past 20 years. This is the kind of research I support!

submitted 20 hours ago • 1 comment

This is a very interesting result. My opinion: RCTs remain one of the more robust methods for p-hacking because the results are more verifiable due to data availability, pre-registration, and simplicity of analyses. In contrast, IV, structural, and simulations are much more flexible for p-hacking.

submitted 1 day ago • 1 comment

I still find it wild that using ‘novel’, private data is consider a huge boost to the credibility of a paper. In fact, the prior should be to discount the results more aggressively because it’s harder to replicate.

submitted 1 day ago • 0 comments

New Workshop alert 🚨 Retrieving and Generating Data using LLMs in Python Join me on May 8th for an intensive workshop that turns messy big data into structured economic data using LLMs. Your registration donation supports students & those in need! #StandWithUkraine

submitted 1 day ago • 2 comments

In light of this result it's perhaps noteworthy that this is the 10th anniversary of the publication of the leading health economics journals' joint manifesto on so-called negative findings. onlinelibrary.wiley.com/doi/10.1002/...

submitted 1 day ago • 1 comment

Will just add to point about the decline in reporting null results: some evidence (in econ) that this problem is concentrated high-ranking journals (top-5s) and not elsewhere. Within sample of all registered RCTs on AEA site (2013-2016), we found reporting nulls in abstract. . .

submitted 1 day ago • 1 comment

As a coauthor on the AEJ:AE report, it’s a lot. Oh my god, It’s a lot. In that paper, outcomes are inconsistently handled both in the code and in the field, the paper’s data is connected to a bunch of other experiments, and we find irregularities in the raw survey files.

submitted 2 days ago • 0 comments

While the (quasi-) experimental methods increase credibility of causal results, there are two related concerns that make studies hard to replicate: 1. decline in reporting of null results 2. increase in use of private sector data a mini thread 🧵based on work with @trfetzer.com

submitted 2 days ago • 1 comment

New paper! DO SOCIAL MEDIA PLATFORMS INCREASE WELL-BEING? THREE PUZZLES. @cfcamerer.bsky.social @petersinger.info @jonathanhaidt.bsky.social papers.ssrn.com/sol3/papers....

submitted 3 days ago • 0 comments

preprint updated - www.biorxiv.org/content/10.1... Each of us perceives the world differently. What may underlie such individual differences in perception? Here, we characterize the lateral prefrontal cortex's role in vision using computational models ... 1/ 🧠📈 🧠💻

submitted 6 days ago • 4 comments

@prashantgarg.bsky.social & @trfetzer.com introduce a methodology that uses #LLMs to process and synthesise tens of thousands of #economics papers into a 'causal knowledge graph', which identifies which economic concepts are linked. cepr.org/voxeu/column... #EconSky

submitted 7 days ago • 3 comments

🚀 Exciting Opportunity! 🚀 #AI keeps me up at night. What will happen to work & workers? 🤔 I'm partnering with Schmidt Sciences to offer up to $200,000 in funding for early-career researchers! 💡 🗓️ Apply by March 31 2025 🙏 spread the word! 🔗 Learn more & apply schmidtsciences.org/ai-at-work #EconSky

submitted 7 days ago • 1 comment

Pre-register your priors...

submitted 9 days ago • 1 comment

Guy sends me an email, thanking me for saving him "days" of work. The latest update of the meta-database on the social cost of carbon took me weeks. It's now online: ideas.repec.org/p/arx/papers...

submitted 14 days ago • 0 comments

🚨New Paper 🎻 On Bob Dylan: A Computational Perspective In this work, I use LLM (o3) to explore how Dylan’s ever-evolving lyrical themes break expectations and challenge convention (i.e., dishabituation). This method can be used an artist’s cultural and creative evolution in general.

submitted 21 days ago • 3 comments

New on the SSPP: What factors predict publication in a Top 5 economics journal? We welcome your predictions on a study by @prashantgarg.bsky.social and @trfetzer.com analyzing author and paper characteristics. socialscienceprediction.org/s/ba04az ⏱️ 10 min 📚 Field: Econ 📅 Closes: April 30

submitted 22 days ago • 0 comments

What Predicts a Top 5? @trfetzer.com and I are exploring which factors boost a paper’s chances of landing in top econ journals (AER, QJE, JPE, EMC, ReStud). Feel free to submit your priors! 👉 Take the survey here: socialscienceprediction.org/predict/r/f7... ⏲️5-10 min

submitted 23 days ago • 2 comments

Fantastic paper and dataset: Prevalence of processed foods in major US grocery stores www.nature.com/articles/s43...

submitted 23 days ago • 0 comments