AI and Machine Learning Models in Cross-Asset Class Investment Risk Analysis: A Case Study of Real Estate and Equities Markets

This research explores the application of Mean-Variance Optimization (MVO) and advanced AI and Machine Learning (ML) models in constructing efficient investment portfolios that balance return and risk across real estate and equities markets. Through a comprehensive review of existing literature, we identify key gaps in traditional portfolio optimization methods, such as their sensitivity to input data and limited adaptability to dynamic market conditions. This review underscores the importance of integrating AI-driven techniques to address these limitations, offering improved predictive accuracy, risk management, and sustainability in investment strategies. Our study analyzes historical data from 2010 to 2023 to evaluate the impact of MVO and AI models on portfolio performance, emphasizing strategies like diversification, regular rebalancing, and the integration of ESG factors for responsible investing. The results demonstrate that AI models, such as Deep Reinforcement Learning (DRL) and Long Short-Term Memory (LSTM) networks, significantly outperform traditional methods, achieving up to 29.52% cumulative returns and a Sharpe ratio of 0.98. These findings highlight the transformative potential of AI in cross-asset investment risk analysis while emphasizing the need for regulatory safeguards to ensure market stability.