Advanced Graph Theory and Machine Learning in Financial Portfolio Optimization (Ongoing)

The project aims to provide a more robust and adaptive portfolio optimization model using advanced graph theory methodologies, capable of capturing the complex interdependencies between assets and their changing relationships over time.

• Position: Researcher
• Duration: Jan, 2024 - now
• Supervisor: Prof. Xiang Cheng & Mr. Lui Sheng Jie
• Main Contribution:
  • Conducted extensive literature review of existing portfolio optimization methods.
  • Addressed limitations found in previous research.

This project explores the use of advanced graph theory methodologies in optimizing financial portfolios, diverging from traditional statistical and probabilistic approaches. By representing financial markets as complex networks, where assets are nodes and their correlations as edges, we intend to develop a model that not only evaluates the performance of individual assets but also their interconnectivity within this network. Incorporating machine learning, the project will dynamically adjust these network representations in response to evolving market conditions. The initial phase involves an extensive literature review of existing portfolio optimization methods, followed by the development and implementation of a novel approach combining graph theory and machine learning, addressing limitations found in previous research.