[Backtest] 5-Year US Megacap Portfolio Performance & Asset Allocation Report
📊 5-Year US Megacap Portfolio Backtest Report
This report presents a comprehensive 5-year historical backtest (July 21, 2021, to June 23, 2026) evaluating dynamic monthly-rebalanced asset allocation strategies using PyPortfolioOpt and Riskfolio-Lib. The asset universe consists of top US market capitalization technology stocks.
- Universe: AAPL, AMZN, AVGO, GOOG, GOOGL, META, MSFT, NVDA, TSM (9 top US megacap stocks)
- Rebalancing Frequency: Monthly (Transaction Cost: 0.1% per turnover)
- Lookback Period: 1 year (252 trading days) for covariance and expected returns calculation
- GPU Acceleration: PyTorch CUDA (NVIDIA GeForce RTX 4070 Ti)
📈 Performance Comparison Table
| Strategy / Portfolio | Cumulative Return | CAGR | Max Drawdown (MDD) | Sharpe Ratio |
|---|---|---|---|---|
| 🚀 Max Sharpe (PyPortfolioOpt) | 433.44% | 40.68% | -39.11% | 1.09 |
| 🛡️ Risk Parity (Riskfolio-Lib) | 193.18% | 24.52% | -44.64% | 0.88 |
| ⚖️ Equal Weight (Benchmark) | 234.40% | 27.91% | -46.13% | 0.94 |
| 📉 SPY ETF (Market Benchmark) | 72.69% | 11.78% | -25.36% | 0.62 |
📊 Cumulative Equity Curves (5-Year)
Figure 1: Comparison of dynamic allocation model values over time vs passive benchmarks.
🔍 Key Strategy Takeaways
1. The Dominance of Max Sharpe in Tech Bull Cycles
Max Sharpe (PyPortfolioOpt) generated a spectacular return of 433.44%. The model dynamically captured the massive tech rally (particularly NVDA and AVGO) by concentrating weight on the most capital-efficient assets.
2. The Limitation of Risk Parity in Concentrated Universes
While Risk Parity (Riskfolio-Lib) is celebrated for risk mitigation, in a heavily correlated, technology-only universe, it suffered a drawdown of -44.64% in the 2022 bear market. This highlights that risk parity requires a multi-asset universe (bonds, gold, commodities) to successfully perform diversification.
*This performance report was generated automatically by the MoonShot Backtest System.*
댓글
댓글 쓰기