Regime × News-State Conditional Asset Pricing

Overview

Regime × News-State Conditional Asset Pricing is a Python research project that segments daily equity returns across four buckets defined by market regime (bull/bear) and news-state (active/dormant). The framework estimates state-dependent factor exposures using Fama-French 5 factors, FOMC monetary surprises, earnings SUE, and GDELT news sentiment, with a Streamlit interface for analysis.

Highlights

• Designed a conditional asset pricing framework across market regime and news-state buckets
• Estimated bucket-level factor exposures using Fama-French 5 factors, FOMC monetary surprises, earnings SUE, and GDELT news sentiment
• Identified state-dependent shifts in earnings and macro betas relevant to active investors
• Scaling to the full S&P 500 universe with cross-sectional quintile sorts on regime-conditional beta shifts
• Adding sector loading clusters for systematic factor research