nlp_utilities.py: Utilities for NLP
business_summary_classifier.ipynb: Jupyter notebook for sector classification based on business summaries of the Russell 3000®.
Index constituent and GICS level 1 data sourced from Norgate.
TO-DO
- Check business summary encoding
- Add topic analysis / company clustering
- Train / test on historical constituents
- Use to backfill instrument master feeding US equity strategy backtests