Monte Carlo Portfolio Risk Analysis: A Framework for Stress Testing and Risk Measurement

This repository contains the preprint of the paper titled “Monte Carlo Portfolio Risk Analysis: A Framework for Stress Testing and Risk Measurement.” The work presents a Monte Carlo–based portfolio risk analysis framework that integrates geometric Brownian motion (GBM) scenario generation, empirical correlation modeling, regime-based crash stress scenarios, and short-horizon intraday forecasting. The system is designed to simulate a wide range of possible market outcomes and evaluate portfolio performance under both baseline and stressed conditions. The framework computes a comprehensive set of risk and performance metrics, including Value-at-Risk (VaR), Conditional Value-at-Risk (CVaR), drawdown statistics, and distributional measures, alongside visual analytics such as return distributions and percentile-based outcome bands. It also incorporates backtesting procedures and comparative evaluation across historical, parametric, and simulation-based risk measures to assess model calibration and reliability. The study situates the framework within the broader literature on Monte Carlo risk modeling, stress testing, tail risk measurement, and high-frequency financial forecasting, and discusses the assumptions, limitations, and trade-offs involved in balancing model simplicity, computational efficiency, and realism. This version is shared for academic transparency, reproducibility, and early dissemination. A persistent DOI has been assigned via Zenodo to ensure long-term accessibility and citation. The author retains all rights to the work.