Date of Award

Spring 2024

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Computer Science

Committee Chairperson

Md Amiruzzaman, Ph.D.

Committee Member

Richard Burns, Ph.D.

Committee Member

Ashik Ahmed Bhuiyan, Ph.D.

Abstract

This thesis presents a predicting stock index by analyzing the macroeconomic Factors using Machine Learning to generate market sentiment. The study explores popular US stock index funds like S&P 500 to critical economic indicators like GDP, Unemployment rate, Consumer Price Index, Money Supply, Retail Sales, etc. The economic data is collected from open-sourced datasets from the Federal Reserve, NASDAQ, and news websites. Data is cleaned and transformed to build a narrative to estimate quarterly fund returns. Tree-based algorithms like XGBoost and Random Forest are used for the prediction. Market sentiment is generated using traditional natural language processing methods and attempts to leverage large language models to summarize market sentiment. Overall, the index forecast performance is evaluated by different market cycles and geopolitical events using XGBoost Algorithm.

Share

COinS