Machine learning applied to stock price forecasting

Forecasting financial markets is one of the most challenging problems in finance, due to the inherently non-stationary market behavior and the influence of numerous factors, including economic indicators, political events, investor  sentiment, and unforeseen occurrences. In her dissertation, Htet Htet Htun explores the application of machine learning (ML) models in predicting which stocks will outperform the market index by a specified threshold, with a focus on all S&P 500 stocks during the volatile period (2017 – 2022), which was notably impacted by the COVID-19 pandemic. 

Through rigorous computational analysis, Htun evaluates the performance of ML models in forecasting relative returns across various strategies. Her findings reveal that ML models, based on time series data and technical indicators, outperform random stock selection, thereby challenging conventional market theories, such as the Efficient Market Hypothesis and the Random Walk Hypothesis.

Dissertation: https://hdl.handle.net/11370/78ca8033-dcc2-4735-bb7f-b875e9f2f744