Social Networks and Retail Trading
Fulin Li, Joint Program in Financial Economics PhD student
This paper develops an empirical model of opinion dynamics on social networks, and estimates the model using data from Reddit wallstreetbets (WSB). This framework allows me to quantify the effect of social network communications on asset prices and retail trading. The model is a high-dimensional VAR. To achieve identification, I exploit the observed network connections between users, and impose parameter restrictions that capture the network effect. I also address the issue that users infrequently disclose their opinions on the social network, by assuming that a user’s decision to disclose his opinion is an independent Bernoulli process. I derive an OLS-type estimator under further identifying assumptions.