AI in Finance: The Use of Artificial Intelligence in Portfolio Diversification
2 viewsAbstract
This paper investigates the duality of Artificial Intelligence (AI) in financial portfolio management, focusing on the transition from traditional diversification strategies to AI-enhanced models. While conventional methods like Modern Portfolio Theory (MPT) offer simplicity and historical reliability, they often struggle with the nonlinear complexities of modern markets. This study employs a quantitative comparative research design to evaluate the effectiveness of machine learning algorithms—specifically Hierarchical Risk Parity (HRP)—in diversifying portfolios containing non-traditional assets, including cryptocurrencies (Bitcoin) and fintech equities. Empirical results demonstrate that an AI-enhanced portfolio achieved a superior Sharpe ratio (0.95) compared to a traditional portfolio (0.76), suggesting improved risk-adjusted returns through the identification of hidden correlations. However, the study identifies a critical "overreliance trap," where the misuse of AI leads to market homogenization, increased systemic risk (beta), and short-termism. The paper concludes that AI should be viewed as a sophisticated enhancement to human expertise rather than a total replacement. A hybrid "Human-AI" framework is proposed to maintain market stability, ensure accountability, and achieve long-term diversification goals in an increasingly automated financial landscape.
Keywords
Artificial Intelligence, Fintech, Hierarchical Risk Parity (HRP), Machine Learning, Non-traditional Assets, Portfolio Diversification, Risk Management, Sharpe Ratio, Systematic Risk