arXiv q-fin· Kevin Foley, Jonathan Hartadi, Shivesh Prakash, Charishma Takkallapalli, Swapnil Vatsal·· 2 days agoAI Score62
NoLBERT News Topics and the Cross-Section of Stock Returns: A News Text Prediction Study Avoiding Backtest Look-Ahead Bias
Reading the News Without Hindsight: NoLBERT Topics and the Cross-Section of Stock Returns
AI summary
This arXiv paper proposes a NoLBERT encoder pre-trained only on texts from 1976 to 1995, comparing it with LDA, MiniLM, and FinBERT on 1.26 million news articles from 2012 to 2024 to predict returns of the 1000 largest U.S. stocks.
Source:arXiv q-fin · arxiv.org
Factor InvestingMachine LearningLarge Language Model ApplicationsBacktesting MethodsPaper Research#Papers / Research#Machine Learning#Large Language Model Applications#US Stocks#Factor Investing#Backtesting