Dissertation
Decision making with peer opinions on social media platforms
Doctor of Philosophy (Ph.D.), Drexel University
Jun 2020
DOI:
https://doi.org/10.17918/00001092
Abstract
This dissertation investigates how peer opinions from social media could help decision making for investors, consumers, and regulators. In particular, using state-of-the-art text mining and machine learning techniques, we study two platforms that are augmented with social media functionality: an investment-based platform and an online video gaming and distributing platform. The first essay focuses on the relationship between the opinion divergence transmitted through investment-based social media platforms and stock volatility. For each stock, investors may hold different opinions on its performance characterized by bullish, bearish, and neutral. We develop a metric that measures the opinion divergence among investors and test its power in explaining future stock volatility. The results show that investors' opinion divergence is negatively associated with future return volatility across a variety of holding periods. Moreover, the impact of opinion divergence will become attenuated over time. In the second essay, we study peer influence in video game adoption. The main challenge in the identification of peer influence is the confounding effect of homophily that introduces selection bias: people with similar characteristics and preferences are more likely to become friends. Taking advantage of the state-of-the-art recommender system algorithms, this paper develops a novel framework to estimate consumer tastes on products which endogenously drive tie formation and use it to control for selection. In line with previous studies, our results indicate that ignoring the homophily will lead to a biased estimate of peer influence. The third essay investigates the various factors influencing the adoption behavior of online video games. The effects of characteristics of both game producers and game consumers are examined. We find games' category spanning and consumers' information advantage impact the likelihood of game adoption. Moreover, we find evidence supporting a moderating role of users' information advantage on the effect of category spanning.
Metrics
39 File views/ downloads
58 Record Views
Details
- Title
- Decision making with peer opinions on social media platforms
- Creators
- Yang Li
- Contributors
- Chuanren Liu (Advisor)Hande Yurttan Benson (Advisor)
- Awarding Institution
- Drexel University
- Degree Awarded
- Doctor of Philosophy (Ph.D.)
- Publisher
- Drexel University; Philadelphia, Pennsylvania
- Number of pages
- ix, 101 pages
- Resource Type
- Dissertation
- Language
- English
- Academic Unit
- Bennett S. LeBow College of Business; Drexel University
- Other Identifier
- 991014695137104721