End-to-End Python implementation of Wu et al.'s (2025) ICAIF'25 paper. It translates unstructured earnings press releases into quantifiable market signals. Implements oLDA topic modeling, Transformer embeddings (BERT/FinBERT/MPNET), GPT-4o interpretability, and rigorous econometric analysis.

This script utilizes the SEC EDGAR API to download HTML copies of SEC filings in an easy-to-use, compliantly polite, and incredibly effective script.

Automated earnings analysis and stock price prediction using LLMs and financial data

📈 Analyze press releases to predict earnings announcement returns using structured data and natural language processing techniques.