Quantihack 2026 — Team Deanos
Discovering a tradeable signal in spurious correlations.
We found that multiplying the z-scores of global coal production and UK road crash casualties produces a composite signal that predicts UK humidity with r = 0.99. This humidity signal then correlates with agricultural commodity prices at r = 0.75.
Quantihack/
├── report.tex # LaTeX research paper
├── Quantihack.pdf # Compiled paper
├── data/
│ ├── world_coal_production/ # Global coal production (1981-2021)
│ ├── road-casualty-data/ # UK road casualties (2005-2020)
│ └── archive (10)/ # UK weather/humidity (2009-2024)
├── plots/
│ ├── wheat_signal_final.png # Signal → Humidity → Agriculture ETF
│ └── zscore_6yr.png # Z-score discovery across windows
└── scripts/
├── full_zscore_v2.py # Main correlation scanner
├── zscore_hunt.py # Z-score product/ratio search
├── plot_6yr.py # 6-year window analysis & plot
└── plot_wheat_final.py # Final signal vs humidity vs DBA plot
| Relationship | r |
|---|---|
| Signal → UK Humidity | 0.99 |
| Signal → DBA Agriculture ETF | 0.75 |
| Signal → Wheat Futures | 0.73 |
| Signal → Corn Futures | 0.73 |
| Signal → Coffee Futures | 0.72 |
| Signal → Sugar Futures | 0.71 |
pip install pandas numpy matplotlib yfinance pyarrow
python scripts/full_zscore_v2.py # Scan for z-score correlations
python scripts/plot_wheat_final.py # Generate the main plot