Institutional Backtesting
Quant Grade
Event-driven and vectorized simulation on real exchange data with realistic fees, slippage, walk-forward validation and Monte Carlo robustness analysis. Past performance is not a guarantee of future results.
Python 3.13
Backtrader
VectorBT
CCXT
Monte Carlo bootstrap
Simulation configuration
Pipeline: Live fetch → resample → feature build → Backtrader event loop → walk-forward CV → Monte Carlo (2,000 paths) → risk report
Equity curves — strategy comparison(sample — run to load live)
Monte Carlo fan — AI Ensemble (2,000 paths)
Monte Carlo risk report
Median terminal return—
5th percentile (worst)—
95th percentile (best)—
Probability of profit—
Expected Max DD (P95)—
CVaR 95% (tail loss)—
Terminal return distribution
Strategy performance — institutional metrics
AI Ensemble
Ensemble
Return
+42.4%
Max DD
-8.2%
Sharpe
2.14
Sortino
3.06
Calmar
5.17
Win rate
63.2%
Profit factor
2.41
VaR 95%
-2.8%
Trades
34
Buy & Hold
Return
+18.6%
Max DD
-22.4%
Sharpe
0.88
Sortino
1.02
Calmar
0.83
Win rate
100%
Profit factor
1
VaR 95%
-6.4%
Trades
1
RSI Mean-Rev
Return
+26.1%
Max DD
-14.2%
Sharpe
1.42
Sortino
1.88
Calmar
1.84
Win rate
54.1%
Profit factor
1.62
VaR 95%
-3.9%
Trades
48
MACD Trend
Return
+14.8%
Max DD
-16.8%
Sharpe
1.02
Sortino
1.31
Calmar
0.88
Win rate
48.7%
Profit factor
1.28
VaR 95%
-4.6%
Trades
22
EMA Cross
Return
+31.2%
Max DD
-11.4%
Sharpe
1.68
Sortino
2.24
Calmar
2.74
Win rate
57.4%
Profit factor
1.94
VaR 95%
-3.2%
Trades
28
Risk disclosure. Backtest and Monte Carlo results are hypothetical simulations based on historical exchange market data. They include modeled fees (0.10%) and slippage (0.05%) but cannot fully replicate live market conditions such as liquidity gaps, exchange outages, or regime shifts. No forecast or simulation guarantees future performance.