Risk Analysis using Monte Carlo Simulation and Causal Inference in Stock Market

The unpredictability of financial markets poses significant challenges for investors and decision-makers, particularly in forecasting stock price movements and assessing risks. Traditional risk management and forecasting approaches often fail to capture complex market dynamics, including non-linear relationships, macroeconomic factors, and external shocks, resulting in suboptimal decisions and greater risk exposure. A new combined methodology which merges Monte Carlo simulation with causal inference facilitates simulation on a large space of hypothetical scenarios and could eventually discern causal relationships amongst the involved variables, presenting end users with a more explicit knowledge on potential risks. Through such methodologies, a wider variety of users can experience easy application and visual output that they can easily appreciate from very critical data regarding an object of interest. It consists of risk factor identification; estimating future outcomes to quantify extreme values in addition to quantities, such as Value-at-Risk (VaR); and quantifying each identified risk factor’s contribution towards actual outcome. Causal inference methods improve the accuracy of predictive performance by identifying the determinants of price movement through use of the Granger causality test. The framework also proposes mitigation strategies, such as optimal portfolio allocation and stress testing, to reduce risks affecting users’ investment goals. Performance evaluations show the model’s effectiveness, surpassing baseline models with improvements of 63.36% in weekly tests, 82.68% during COVID stress tests and 89.67% in yearly tests. This holistic approach equips decision-makers with the tools to mitigate critical risks, visualize potential outcomes, and align investment strategies with their goals through clear, actionable data.