Big Data Analysis of Volatility Spillovers of Brands across Social Media and Stock Market Performance

Abstract

Volatility is an important metric of financial performance, indicating uncertainty or risk. So, predicting and managing volatility is of interest to both company managers and investors. This study investigates whether volatility in user-generated content (UGC) can spill over to volatility in stock returns and vice versa. Sources for user-generated content include tweets, blog posts, and Google searches. The authors test the presence of these spillover effects by a multivariate GARCH model. Further, the authors use multivariate regressions to reveal which type of company-related events increase volatility in user-generated content.
Results for two studies in different markets show significant volatility spillovers between the growth rates of user-generated content and stock returns. Further, specific marketing events drive the volatility in user-generated content. In particular, new product launches significantly increase the volatility in the growth rates of user-generated content, which in turn can spill over to volatility in stock returns. Moreover, the spillover effects differ in sign depending on the valence of the user-generated content in Twitter. The authors discuss the managerial implications.

Keywords: user-generated content, stock market performance, volatility, multivariate GARCH model, spillover effects, natural language processing.