Downloads · 30 days
0
Tjgguy12/forex-6pair-1d-quant
forex-6pair-1d-quant is a machine learning model from Tjgguy12. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for sklearn. The card lists the license as apache-2.0.
Edges: EURUSD mean reversion, GBPUSD vol clustering, USDJPY vol+carry, AUDUSD mean reversion, USDCAD oil beta. Tail risk USDTRY/ZAR excluded. Do NOT trade USDHKD/EURDKK.
Downloads · 30 days
0
Access
Public
Updated Sep 10, 2026
Repo size
2 MB
Likes
0
Public
Click a slice to open those files.
.pkl2 MB · 99%
From the Hugging Face model README
Edges: EURUSD mean reversion, GBPUSD vol clustering, USDJPY vol+carry, AUDUSD mean reversion, USDCAD oil beta. Tail risk USDTRY/ZAR excluded. Do NOT trade USDHKD/EURDKK.
Training: 5y yfinance (6150 rows, 1300 per pair), Voting XGB+LGBM, 3 folds, 1d embargo. Result: Mean IC -0.012, Ann -6.94% to -1.89% — negative (worse than random). Status: Experimental, NOT production. Needs longer history (2004 Kaggle daily) + proper session features + FinBERT.
Live test 2026-09-09:
Kaggle: https://www.kaggle.com/code/maxvwede/forex-6pair-1d-quant-eur-gbp-jpy Portfolio: 35/25/20/10/10 weighted daily rets, 1bp spread.
Next: Retrain with 2004-2026 Kaggle daily (20y) + London/Tokyo session + oil beta for USDCAD.