Fast-forward through 2008, COVID, demonetisation. Watch a real ₹1,00,000 portfolio move in real time, then see where our ML model got it wrong.
Illustrative sector snapshot (design reference values, not a live per-sector feed).
Anchored to publicly reported Sensex/gold milestones. This is a teaching simulation, not a live market feed.
*A 5-day-lagged view of the real actual line, calibrated to the real MAE reported alongside, not literal day-by-day model output. Full breakdown on crash replay.
Random Forest and XGBoost's directional accuracy, which beats chance in calm 2017 conditions, flips to significantly worse than random during COVID (p < 0.01). A real result from this project's own evaluation, shown here rather than hidden.
Open ML Lab →