Emil Palikot Susan Athey Keshav Ayush Kanodia CAREER utilizes transformer-based techniques for job sequence prediction, combining large-scale resume data with detailed survey datasets to improve the accuracy of economic predictions. arxiv.org/abs/2202.08370

🚨Thrilled to share my JMP with #EconTwitter! 🚨 "The Personalization Paradox: Welfare Effects of Personalized Recommendations in Two-Sided Digital Markets" Here is a short thread🧵 (1/n) 🔗 aaron-kaye.com/home

Great paper with depressing result: 1. Recommender systems help consumers find better-matching products (yay!) 2. Sellers use the better matches to raise prices (boo!) What's the overall effect (data from Expedia): welfare loss for consumers!!! apkaye.github.io/aaronpkaye.web…

My PhD Syllabus for Introduction to Quantitative Marketing Rotman School. Updated for 2024. Comments welcome. avigoldfarb.com/s/Syllabus_RSM…

Rose Bowl champs! 〽️ #GoBlue

Susan Athey in her Presidential address then describes her surprise in 2008 learning about the hundreds of A/B tests run in tech companies. She then described how AB testing is only the first step in the innovation process

How to know which units have the higher treatment effect? Hard to tell ex ante. In exp w ideas42, is it better to target individuals more, or less, likely to fill FAFSA with a nudge? In my term, who has the higher persuasio rate? Turns out the people MORE likely have higher TE

Curious about AI/ML and social impact? Faculty Director Susan Athey discussed some of the exciting work happening at Golub Capital Social Impact Lab and her insights on future research in this interview before her #ASSA2024 Presidential Address. ow.ly/o8e650Qq7aZ

Marketing Science Greg Lewis One insight coming out of this paper: ranking based on utilities may not maximize welfare when consumers engage in search. Instead, you want to promote products that deliver the best surprise relative to what consumers expect pre-search.

Customized Content for Learning: Susan Athey Keshav Emil Palikot Ayush Kanodia built a personalized recommendation system for stories ("Netflix for stories!") on a reading app in India. Students used it more and read more when they did! x.com/gsbsilab/statu… 7/11

An additional comment I would add is that I view this as one of a number of developments indicative of a decline in the economics profession's standing in the policy making process and in the operations of the government. The development and maintenance of robust statistical

📣ICYMI: "CAREER: A Foundation Model for Labor Sequence Data" is now published in Transactions on Machine Learning Research! Keyon Vafa Emil Palikot Tianyu Du Ayush Kanodia Susan Athey Blei Lab

Heading into the job market this year with a field experiment in economics, business, or policy? Consider submitting your job market paper's abstract to #AFE2024. It's an excellent opportunity to gain insightful feedback and connect with researchers across various fields

Folks, the Behavioral IO and Marketing conference is in two weeks. You can still register. Great talks and fun times await. Join us.

Grateful to receive the Best Paper Award at the Doctoral Workshop on the Economics of Digitization! It was fantastic meeting everyone and learning about their impressive projects. Thank you to the University of Warsaw, CESifo, and all the organizers! linkedin.com/posts/oliver-f…

Results from the OpenResearch RCT are out: 1000 low income participants randomly assigned to receive $1000/mo for 3 years. I’ll do a thread on some specific results shortly, but I just want to say how proud I am to be part of this study and highlight what we’ve accomplished (1/n)

Aaron Kaye presenting some 🔥research on the equilibrium (unexpected) effects of personalization in recommendation systems in two sided digital platforms. Stanford University #SITE #MarketDesign Work shows price adjustment by supply side can lead to welfare loss even though rec's

Excited to announce a new paper on dark patterns and competition online! You’ve seen cookie banners all over the internet, some with dark patterns that make it hard to reject cookies. We conducted a field experiment about these banners to study how people choose to share their