Portfolio selection is an important part of fund management as it contributes to investors’ economic growth. Investing is investing money to obtain an additional or specific advantage over money. Investment involves not only profit (return), but also risk that the investor bears. The higher the return an investor expects, the higher the risk the investor takes. Proper portfolio construction can minimize the level of risk to the expected value of an individual stock portfolio. Equity portfolio optimization, therefore, plays an important role in setting an investor’s investment portfolio strategy. In recent decades, there have been great advances in financial mathematics.
By harnessing the increasingly powerful computational power, these problems have been addressed in new ways, developing new algorithms to trade, model, and forecast mostly automatedly. With the recent increasing adoption of machine learning approaches, genetic algorithms have emerged as one of the most widely used stochastic optimization approaches for solving complex optimization problems. The proposed approach aims to develop a machine-learning solution to simplify investment decision-making by rapidly generating optimal investment portfolios. The suggested study will train a genetic algorithm to help investors select assets with the greatest return. The proposed solution is a Genetic Algorithm- based model that: selects K assets from the universally available assets, includes them in the portfolio, and the capital weights to invest in to minimize risk and maximize portfolio returns
/ Allocate shares.
The study exploited 391,108 assets to expound on the utilization of genetic algorithms in constructing optimal investment portfolio and experimented two optimization approaches, that is single objective case and multi objective approach. The maximization of Sharpe ratio resulted to a high-risk portfolio with risk rising to 49.24% during asset selection step and 787.31% capital allocation while the multi objective approach yield was dependent on the objective weights. For risk seeking investors who would rather miss their risk goal than missing their return goal, the strategy allocates high amounts to risky assets. As the investor becomes more risk averse, the strategy allocates high capital proportion to less risk assets.
Professional Studies (MS)
Department, Program, or Center
Graduate Programs & Research (Dubai)
Khalil Al Hussaeni
Al Ali, Maythaa and Al Dossari, Shaikha, "Customised Investment Optimization Using Genetic Algorithms" (2023). Thesis. Rochester Institute of Technology. Accessed from