🚀 MAJOR RELEASE The TRACE Corporate Bond Data Pipeline — October 2025 code release is live! Part of the Open Bond Asset Pricing initiative. Our goal: corporate bond data is controversial. We give researchers complete control over their TRACE corporate bond data cleaning procedures with a comprehensive, customizable code pipeline written in Python. Build your corporate bond database with the filtering choices you think are reasonable and defensible. What's included: ✅ Full user control — Choose from extensive filtering options: trading volume thresholds, bond types, time-of-day restrictions, error correction parameters - if you don't agree on a data filter - simply exclude it, or augment it! ✅ Automated processing — Transform raw intraday TRACE transactions into clean daily pricing panels for Enhanced, Standard, and 144A TRACE datasets ✅ Error correction algorithms — Don't just throw data away! Fix it first with the decimal-shift corrector (someone fat fingered a trade of $100 as $0.01 -- fix it!) and bounce-back algorithms that automatically detect, correct and eliminate transaction price errors ✅ Transparency — Generates detailed 400+ page reports documenting every filter decision and every bond that is affected by the error corrector/eliminator algorithms ✅ Customization — Adjust parameters to match your research needs and create defensible, reproducible datasets that you have full control over 🚀 Stage 0 of the pipeline is in public beta and fully open-source. We're actively seeking researchers to test the code, contribute improvements, and help us build a community-driven corporate bond database. Stage 1 and 2 are coming end of November. All code is publicly available and ready to use: github.com/Alexander-M-Dicke… Please give it a star! ✨ 🔗 Questions, suggestions, or want to contribute? Contact [email protected] Thanks to Giulio Rossetti and Mengshi Jia for providing amazing feedback.