Stochastic Portfolio Engine — Python

Overview

A comprehensive quantitative finance notebook covering portfolio construction, CAPM analysis, Value-at-Risk, and Monte Carlo simulation applied to SBRY.L (Sainsbury's) and TSCO.L (Tesco) benchmarked against the FTSE 100 (^FTSE). Data spans May 2024 – May 2025 sourced via yfinance.

The notebook follows a structured step-by-step approach from data loading through to optimal portfolio identification.

---

What's Inside

Step 1: Data Loading & EDA

• Downloads price data for SBRY.L, TSCO.L, ^FTSE using yfinance

• • Computes descriptive statistics and inspects data quality (269 trading days)

  • Step 2: Return & Excess Return Calculation

  • • Computes daily log/simple returns

    • • Calculates excess returns over a 4% annual risk-free rate (FRED-based): rf_annual = 0.04

      • Step 3: Max Drawdown Analysis

      • • SBRY max drawdown: -23.67%

        • • TSCO max drawdown: -20.76%

          • Step 4: Value at Risk (VaR)

          • Historical quantile-based VaR at two confidence levels:

          • • VaR 95%: SBRY -2.32%, TSCO -1.68%

            • • VaR 99%: SBRY -4.16%, TSCO -4.00%

              • Step 5: CAPM Analysis

              • OLS regression of excess asset returns against FTSE excess returns:

              • • Yields beta (systematic risk) and alpha (abnormal return) for each stock

                • Step 6: Monte Carlo Simulation (1,000 portfolios)

                • • Generates random Dirichlet-weighted portfolios over SBRY.L and TSCO.L
                  • Computes expected return, volatility, and Sharpe ratio for each simulation

                  • Identifies Best Sharpe, Worst Sharpe, and Lowest Risk portfolios

                  • Best Sharpe Result:

                  • Return: 18.4% | Volatility: 20.1% | Sharpe: 0.938

                  • Weights: SBRY 0.4% / TSCO 99.6%

                  • Lowest Risk Result:

                  • Return: 18.4% | Volatility: 20.1% | Sharpe: 0.715

                  • Weights: SBRY 34.2% / TSCO 65.8%

                  ---

                  Key Results

                  Metric	SBRY.L	TSCO.L
                  Expected Return (ann.)	7.87%	23.81%
                  Volatility (ann.)	23.46%	21.07%
                  VaR 95%	-2.32%	-1.68%
                  VaR 99%	-4.16%	-4.00%
                  Max Drawdown	-23.67%	-20.76%

                  ---

                  Technologies Used

                  Tool	Purpose
                  Python 3.8+	Core language
                  yfinance	Market data retrieval
                  NumPy / Pandas	Data manipulation
                  SciPy (stats, linregress)	CAPM regression
                  Matplotlib / Seaborn	Visualisations (skew, kurtosis)
                  Google Colab	Execution environment

                  Getting Started

                  git clone https://github.com/srhimal149/Stochastic-Portfolio-Engine-Python-.git
                  cd Stochastic-Portfolio-Engine-Python-
                  pip install yfinance numpy pandas scipy matplotlib seaborn jupyter
                  jupyter notebook

                  Author

                  Md Saifur Rahman Himal MSc FinTech (Distinction Candidate) | Quantitative Finance & ML LinkedIn | GitHub

                  License

                  MIT License