Essays on Social Media in Asset Pricing
Open AccessThis dissertation contains three interdependent parts. The first part uses a novel datasetof corporate tweets to investigate Twitter usage by the US publicly traded firms. The analysis implements six machine learning algorithms to classify tweets and uses Natural Language Processing (NLP) to extract information and sentiment scores from tweets. This methodology takes advantage of Twitter-specific numerical measures to enhance prediction accuracy. The second part is a joint research project that introduces the idea that social media has become a source of non-diversifiable risk. The findings suggest that companies more exposed to social media risk are more likely to be on social media networks and tend to earn higher returns. Most of the return differential represents compensation for non-diversifiable social media risk. Notably, companies that currently have higher social media betas and earn higher returns, did not earn higher returns 20 years ago, before the era of social media. The third part extends the results and analyses of the first two sections. It explores the informational environment in capital asset markets and investigates how information is incorporated in asset prices and returns. The analysis takes advantage of the complexities in the interaction between the company-controlled and public-controlled information flow on Twitter to reveal implications relevant to systematic and idiosyncratic risk and capital assets pricing. The findings show that firms’ presence on Twitter can enhance brand loyalty and stability of future cash flows and reduce idiosyncratic risk. The research shows that while an increased firm’s Twitter activity tends to increase uncertainty due to the initial difficulty of interpretation, this uncertainty tends to be resolved within approximately one month.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.
| Thumbnail | Title | Date Uploaded | Visibility | Actions |
|---|---|---|---|---|
|
|
Hosseini_gwu_0075A_15518.pdf | 2021-05-10 | Open Access |
|