Portfolio Optimization & Risk Analysis using Monte Carlo Simulation

Project Overview

This project is a comprehensive Python-based implementation of Modern Portfolio Theory (MPT). The primary goal is to analyze a portfolio of financial assets (stocks, ETFs) and identify optimal asset allocation.

The tool fetches historical market data, performs exploratory data analysis (EDA), calculates risk metrics (including Value at Risk - VaR), and then runs 10,000+ Monte Carlo simulations to map the Efficient Frontier. Finally, it identifies the portfolio with the maximum Sharpe Ratio (best risk-adjusted return) and the minimum volatility portfolio.

Key Features

• Dynamic Data Ingestion: Fetches historical price data for any list of tickers using the yfinance library.
• Data Cleaning: Intelligently handles NaN values resulting from different exchange holidays (e.g., WSE vs. NYSE) by forward-filling (ffill) prices and filling 0 for volume.
• Exploratory Data Analysis (EDA): Calculates daily returns, analyzes return distributions (histograms), and visualizes the correlation matrix (seaborn.heatmap) to understand diversification benefits.
• Risk Modeling: Implements historical Value at Risk (VaR) to quantify potential losses.
• Monte Carlo Simulation: Runs over 10,000 simulations with random portfolio weights to generate the Efficient Frontier.
• Optimization: Identifies and highlights two key portfolios:
  1. Minimum Volatility Portfolio (The "safest" option).
  2. Maximum Sharpe Ratio Portfolio (The "optimal" risk-adjusted option).

Core Insights

• The correlation analysis confirmed (e.g., a low correlation between Polish WIG20 assets and US tech stocks), demonstrating the clear benefits of geographic diversification.
• The generated Efficient Frontier clearly illustrates that achieving higher expected returns requires taking on exponentially higher levels of risk (volatility).
• The Max Sharpe Ratio (optimal) portfolio is not the same as the Minimum Volatility (safest) portfolio, presenting a clear trade-off for different investment strategies.

Tech Stack & Libraries

• Python 3.12
• Pandas: For data manipulation, time-series analysis, and DataFrame management.
• NumPy: For high-performance numerical computing and matrix operations (dot products, covariance).
• Matplotlib & Seaborn: For static data visualization (histograms, heatmaps, density plots).
• yfinance: For fetching historical market data from the Yahoo Finance API.
• Git & GitHub: For version control.