Kecerdasan Genetic Algorithm sebagai Inovasi dalam Optimasi Portofolio Saham di India: Analisis Komparatif dengan Model Markowitz

الملخص

This study aims to analyze and compare the performance of stock portfolio optimization using the Markowitz model and Genetic Algorithm (GA) in the Indian stock market. The research is motivated by the high volatility of the Indian stock market and the limitations of the mean-variance approach in dealing with dynamic market conditions. This study uses the foundation of Modern Portfolio Theory (MPT), the concept of efficient frontier, and evaluation of portfolio performance through Sharpe, Sortino, and Omega ratios. The research method used is a comparative quantitative approach with secondary data in the form of stocks that are members of the Nifty 50 and S&P BSE TASIS Shariah 50 indices for the period January 1, 2022 to October 30, 2025. Data processing is done using Python with the help of statistics and optimization libraries. The results show that the combined portfolio produces the best performance with an expected return of 2.71%, risk of 0.0421, Sharpe Ratio of 0.6423, Sortino Ratio of 1.6096, and Omega Ratio of 5.0821. Markowitz’s model generally results in a slightly more optimal efficient frontier than GA, although GA still exhibits competitive and flexible performance in handling complex optimizations. This study concludes that diversification across stock groups is able to increase risk-return efficiency and provide alternative investment strategies according to the investor’s risk profile.