The Impact of Social Media Sentiment on Stock Market Based on User Classification

Abstract

The relationship between social media sentiment and the stock market has been receiving much attention. Based on the perspective of social media user classification, using Weibo and Shanghai Composite Index data, quantile regression and instrumental variable quantile regression (IVQR) model is built to explore the impact of sentiment and sentiment fluctuation on authenticated and non-authenticated users on stock market returns respectively. The research results show that Weibo sentiment and sentiment fluctuation have a positive impact on stock market returns, but the effects of the two types of users are different. The sentiment of authenticated users has a stronger and longer impact on stock market returns, while only moderate sentiment and sentiment fluctuation of non-authenticated users have a positive impact on stock market returns. The research provides evidence for the relationship between social media sentiment and stock market, and has some practical significance for both social media platforms and users.