polyhedge

maps a stock portfolio's implicit macro bets to live polymarket prediction markets and sizes hedges where tradfi diverges from the crowd. yale hack 2025.

stack

layer	tools
market data	yfinance, fredapi
factor regression	statsmodels ols
prediction markets	polymarket rest api
numerics	numpy, pandas, scipy, scikit learn
frontend	typescript

pipeline

1. regress each holding against 18 macro factors to get per factor betas
2. dollar weight betas into one portfolio factor profile
3. map factors to polymarket categories, derive a tradfi implied probability
4. classify the gap vs market price: under 8pp noise, 8 to 15pp hedge, over 15pp strong
5. size with quarter kelly, 3% single position cap, 10% total

run

pip install -r requirements.txt
export FRED_API_KEY=...          # free at fred.stlouisfed.org
python3 -m polyhedge.pipeline --portfolio portfolio_example.json

outputs json with per market divergence (tradfi implied minus polymarket price), effective_edge (confidence discounted, used for sizing), and insurance_cost_pct.