Analyzing Portfolio Risk and Performance Using Python: A Deep Dive into Financial Metrics

📊 Introduction

In today’s dynamic financial markets, understanding the risk and performance of a stock portfolio is more crucial than ever. As part of my ongoing exploration into quantitative finance and Python-driven analytics, I developed a Jupyter Notebook-based project to perform a comprehensive risk-return analysis on a portfolio consisting of:

• Mastercard (MA)
• Visa (V)
• Coinbase (COIN)
• Robinhood (HOOD)

This project bridges theory with practical implementation using Python, allowing for a robust analysis of returns, volatility, risk-adjusted performance, and tail-risk metrics.

🛠 Tools & Libraries Used

• Python (via Jupyter Notebook on Anaconda)
• NumPy, Pandas – data wrangling and calculations
• Matplotlib, plotly – data visualization
• yFinance – historical market data
• SciPy – advanced statistical measures

📈 Key Performance & Risk Metrics

The analysis dives deep into the following quantitative metrics:

🔹 Return Measures

• Simple Returns – basic percentage returns
• Log Returns – preferred for continuous compounding and portfolio analysis

🔹 Risk Measures

• Volatility – standard deviation of returns, capturing price fluctuations
• Maximum Drawdown – largest observed loss from a peak to a trough

🔹 Risk-Adjusted Performance

• Sharpe Ratio – returns adjusted for total risk
• Sortino Ratio – returns adjusted for downside risk only
• Treynor Ratio – returns per unit of market risk (Beta)
• Calmar Ratio – returns relative to drawdowns

🔹 Market Sensitivity

• Alpha – excess returns beyond the benchmark
• Beta – sensitivity to market movements (systematic risk)

🔹 Tail-Risk Metrics

To understand worst-case scenarios, I calculated:

• Value at Risk (VaR) at 90%, 95%, and 99% confidence levels
• Conditional VaR (Expected Shortfall) at the same levels, which quantifies the average loss given that the VaR threshold has been breached

📉 Insights Derived

• Coinbase and Robinhood, while offering higher upside, significantly increased the portfolio's volatility and drawdown potential.
• Visa and Mastercard provided stability and strong Sharpe/Sortino ratios, indicating strong returns per unit of risk.
• Tail-risk measures like Conditional VaR revealed potential hidden dangers in high-volatility assets that standard deviation alone cannot show.

The portfolio-level analysis allowed for a balanced interpretation of risk vs return, revealing trade-offs between diversification, downside protection, and market exposure.

📂 GitHub Repository

🔗 View the complete code and notebook on GitHub: 👉 https://github.com/hardiktrehan1/portfolio-risk-analysis

You can clone it, run it, and even tweak the weights or assets to suit your own portfolio strategy.

💬 Conclusion

This project is a testament to the power of data-driven decision making in finance. Whether you’re a quant analyst, a portfolio manager, or an investing enthusiast, understanding the true risk-return profile of a portfolio is essential for long-term success.