Date of Graduation

5-2023

Document Type

Thesis

Degree Name

Bachelor of Science in Business Administration

Degree Level

Undergraduate

Department

Finance

Advisor/Mentor

Santamaria, Sergio

Abstract

Many institutions, investors, and researchers have sought to gain insights into consumer attitudes and behaviors by leveraging social media. However, trading strategies utilizing social sentiment have been observed to often provide returns roughly equivalent to or less than those of standard market indexes. This observation raises several questions: does the efficacy of social sentiment analysis vary from firm to firm? Are price or volume directly correlated with social sentiment? Does social sentiment display any predictive ability? The following analysis attempted to answer these questions by comparing one week’s worth of data from Twitter users to price and volume data from six publicly traded companies. Although more data is needed to validate this analysis’ findings, there was evidence to suggest that social sentiment’s efficacy varies significantly between firms, changes in prices and volume do not seem to be correlated to changes in social sentiment, and there does appear to be some patterns that could help predict trend reversals in price.

Keywords

social media, finance, stock market, sentiment, Twitter

Share

COinS