We trained XGBoost on 18 months of Bank Nifty data: features atm_iv, pcr_strikes, oi_buildup, vix_india, max_pain_distance, all lagged one bar. Walk-forward with 3-month train, 1-month test, rolled 12 times. Out-of-sample accuracy held 58-61 percent versus 54 percent naive, with Sharpe near half the in-sample.
The filter (band 0.58-0.80, VIX z<2, max-pain>0.3%, DTE>1) cut drawdown by excluding panic-regime trades. This is a research result, not a live track record; validate on your own data before risking capital.
Q: What was the out-of-sample accuracy?
A: 58-61 percent across 12 walk-forward folds, vs 54 percent naive.
Q: Did the filter help?
A: Yes — it cut drawdown by excluding high-VIX panic trades.
Q: Is this a live result?
A: No, research only; validate before using real capital.
📞 Free Nifty Options AI help & resources — WhatsApp: 9169650895
By Shakti Tiwari · Options AI research pillar. NISM XII certified. Educational only, not investment advice; verify before acting.