Predicting Stock Prices and Optimizing Portfolios: A Random Forest and Monte Carlo-Based Approach Using NASDAQ-100

In recent years, machine learning has gained substantial traction in financial markets, particularly in predicting stock prices and optimizing investment portfolios. Traditional methods for stock prediction, such as fundamental and technical analysis, have limitations in capturing complex market patterns. This study explores the application of the Random Forest model in stock price prediction and portfolio optimization using NASDAQ-100 constituent stocks. By combining return predictions from the Random Forest model with Monte Carlo simulations for portfolio construction, the research aims to create portfolios that maximize returns while maintaining controlled risk levels. The results indicate that the constructed portfolios significantly outperformed the NASDAQ-100 benchmark in annualized returns, though they exhibited higher volatility and risk, particularly during market downturns. While the machine learning approach performed well in normal conditions, certain limitations became evident during extreme market environments. Future research could address these issues by incorporating broader diversification and more advanced risk management techniques to enhance portfolio stability.