Author ORCID Identifier

https://orcid.org/0009-0007-6034-9158

Date of Award

5-3-2024

Degree Type

Dissertation

Degree Name

Doctor of Business Administration (DBA)

Department

Computer Information Systems

First Advisor

Dr. Likoebe Maruping

Second Advisor

Dr. Po-An Hsieh

Third Advisor

Dr. Denish Shah

Abstract

For many companies, some consumers enthusiastically follow brands and may have insights rivaling those of professional financial analysts based on their knowledge of the companies. Often, these brand-loyal consumers express their thoughts and opinions on social media may be received by others in their communities, driving users to follow them based on the perception of expertise and trustworthiness. In academic literature, such users are referred to as Social Media Influencers (SMI).

The present study investigates the relationship between sentiments of Twitter posts and abnormal stock price returns. It further explores if source credibility operationalized as followership affects this relationship. SMIs are perceived to have higher source credibility, and it is expected that the relationship is stronger with SMIs than with non-SMI users. Sentiment analysis categorizes tweets into positive, neutral, and negative based on pre-trained models and machine learning.

Tweets made between 2017 and 2022 for four specific firms are analyzed using an event study approach. Events are identified using an automated algorithm, and abnormal returns are estimated using the Market model. Tweets are split based on percentile rankings of posting authors’ follower counts.

The study finds that positive sentiments are generally statistically significant in identifying positive cumulative abnormal returns. Furthermore, the novel approach of gradually including fractiles of follower numbers shows that the significance of abnormal returns is not homogenous across all users. Contrary to expectations, statistical significance is stronger for a longer duration around identified events with tweets posted by users at the bottom 30% of followership. In contrast, tweets made by the users with more followers are not statistically significant until 80%. This finding suggests that the sentiments of tweets from users with a lower number of followers are more strongly related to the abnormal returns of stock prices.

This study shifts the focus of most extant research by using a broader set of tweets from “ordinary” users instead of investment-oriented users in exploring the relationship. The study also contributes to the relatively underexplored effect of SMI by taking an iterative approach to studying tweets through comparative analysis across different numbers of followers.

DOI

https://doi.org/10.57709/36943015

File Upload Confirmation

1

Download

Share

COinS