📈 Can a Data-Driven Portfolio Beat the Market?

This project explores a quantitative approach to portfolio construction by leveraging volatility (VIX), sector rotation, and regression analysis. We test whether a dynamically weighted strategy can outperform the S&P 500 over a five-year period.

📌 Objective

• Build a sector ETF portfolio influenced by VIX trends and sector correlations.
• Use CAPM Beta, log returns, and regression to adjust weights.
• Back-test performance against SPY from 2019 to 2024.

📁 Dataset

• Simulated daily log returns for:
  • SPY (S&P 500)
  • VIXY (VIX proxy)
  • 8 Sector ETFs (XLK, XLF, XLY, XLV, XLE, XLI, XLB, XLRE)

🧪 Methodology

• Calculate daily log returns and rolling volatility.
• Compute CAPM Beta (60-day) of each sector vs SPY.
• Generate VIX trend signal to allocate weights dynamically.
• Normalize weights and compute portfolio returns daily.
• Visualize cumulative returns vs S&P 500.

📊 Results

The Quant Portfolio achieved ~10% higher cumulative return than SPY over the 5-year period.

📂 Files

• DataDriven_Portfolio_Backtest.xlsx – Cleaned dataset and calculations
• Quant_Portfolio_vs_SP500.png – Strategy performance chart
• DataDriven_Portfolio_Report_Mayank_Agarwal.docx – Final report

📘 Conclusion

Volatility-informed regression models provide alpha opportunities. This strategy showcases the value of macro indicators like VIX in active portfolio management.

---

👤 Author: Mayank Agarwal
📧 Connect on LinkedIn