ABSTRACT
The authors use crawler software to capture the data of investor post content and reading volume from the stock bar of Eastmoney and then utilize textual analysis method to construct positive and negative investor sentiment indicators. The authors use VAR model to examine the Granger causality between investor sentiment and stock returns for the Chinese stock market. The results indicate that negative sentiment indicators have more explanatory power in determining stock returns than positive sentiment indicators. Evidence from impulse responses also shows that investor sentiment reacts to stock returns in short-term.
Disclosure statement
No potential conflict of interest was reported by the author(s).