One daily newspaper column explains the US stock market better than 𝟭𝟬𝟬+ economic indicators combined. 📰 In 1988, economist Richard Roll showed finance couldn't explain most stock moves, even 𝘸𝘪𝘵𝘩 𝘩𝘪𝘯𝘥𝘴𝘪𝘨𝘩𝘵. Nearly 40 years later, 𝗕𝗿𝘆𝗮𝗻 𝗞𝗲𝗹𝗹𝘆 and co-authors (Y

One daily newspaper column explains the US stock market better than 𝟭𝟬𝟬+ economic indicators combined. 📰 In 1988, economist Richard Roll showed finance couldn't explain most stock moves, even 𝘸𝘪𝘵𝘩 𝘩𝘪𝘯𝘥𝘴𝘪𝘨𝘩𝘵. Nearly 40 years later, 𝗕𝗿𝘆𝗮𝗻 𝗞𝗲𝗹𝗹𝘆 and co-authors (Yale, Johns Hopkins, AQR) revisited the question with AI. 🔍 𝗪𝗵𝗮𝘁 𝘁𝗵𝗲𝘆 𝗱𝗶𝗱 AI read 20+ years of Wall Street Journal market recaps and pulled out the reasons given for each day's move: earnings, the Fed, jobs data, wars. Anything describing price moves was removed, so no peeking at the answer. 📊 𝗪𝗵𝗮𝘁 𝘁𝗵𝗲𝘆 𝗳𝗼𝘂𝗻𝗱 Economic data, company earnings, analyst forecasts: basically 𝟬% Who's buying and selling: about 𝟭𝟬% News reasons: about 𝟯𝟬% of daily moves and 𝟰𝟬% of monthly ones (all tested on data the model never saw) 💡 𝗪𝗵𝘆 𝗶𝘁 𝗺𝗮𝘁𝘁𝗲𝗿𝘀 A jobs number alone says little. "Worse than expected" says a lot. That context lives in words, and AI can now measure it. 🤔 𝗧𝗵𝗲 𝗰𝗮𝘁𝗰𝗵: reporters write these reasons after seeing the move, so they explain it in hindsight but can't prove what caused it. 📄 𝘌𝘹𝘱𝘭𝘢𝘪𝘯𝘪𝘯𝘨 𝘙𝘦𝘢𝘭𝘪𝘻𝘦𝘥 𝘙𝘦𝘵𝘶𝘳𝘯𝘴: 𝘙𝘰𝘭𝘭'𝘴 𝘙² 40 𝘠𝘦𝘢𝘳𝘴 𝘓𝘢𝘵𝘦𝘳 #AI #ArtificialIntelligence #GenerativeAI #LLM #MachineLearning #DataScience #NLP #Finance #Investing #StockMarket #FinancialMarkets #Economics #QuantFinance #AssetManagement #AssetPricing #WallStreet #HedgeFunds #FinTech #Research #Journalism Link to the paper: I started a monthly digest that cuts out all the noise in #Quant Finance & #AI/#LLM research. Just the things that actually matter. Friends keep telling me it saves them a ton of time. Sign up here: