Master’s Project Defense for Elizabeth Cox

Tuesday November 15, 2022, 11:00 am – 1:00 pm in Student Commons Building 4119

Title: Methods in Portfolio Optimization and Their Applications

Committee: Joshua French (advisor), Erin Austin (advisor), Hani Mansour

Abstract: Portfolio optimization, predominantly a financial problem, highlights the fact that investors may only increase their returns by taking on more risk. Portfolio optimization methodologies that were designed for the financial world can also be applied and provide useful insight into any problem in which we wish to select optimal quantities of any goods or services. We compare three predominant methods in portfolio optimization: Markowitz’s mean-variance optimization, Black-Litterman portfolio optimization (a Bayesian model), and a deep learning method. These methods are applied to a series of simulated data and EU energy supply data to understand how performance changes when applied to non-financial data, and if methods respond to existing barriers or physical limitations present in real world applications.