We investigate how online investor sentiment impacts stock risk, measured as Value-at-Risk (VaR). We extrapolate online investor sentiment from information on the stock forum on the 100 constituent stocks of the Shenzhen index using a self-written code to collect daily online postings from 2016 to 2022. Then, we rely on algorithms to classify them. Using quantile regressions and controlling for firm-specific factors and COVID-19, we document that stronger sentiment increases VaR while decreasing VaR on a lagged 7-day horizon. As we move to a longer horizon (20 days), the effect vanishes as more information becomes incorporated into the stock prices.
Is it all about noise? Investor sentiment and risk nexus: evidence from China
Cardillo, Giovanni
;Harasheh, Murad
2023
Abstract
We investigate how online investor sentiment impacts stock risk, measured as Value-at-Risk (VaR). We extrapolate online investor sentiment from information on the stock forum on the 100 constituent stocks of the Shenzhen index using a self-written code to collect daily online postings from 2016 to 2022. Then, we rely on algorithms to classify them. Using quantile regressions and controlling for firm-specific factors and COVID-19, we document that stronger sentiment increases VaR while decreasing VaR on a lagged 7-day horizon. As we move to a longer horizon (20 days), the effect vanishes as more information becomes incorporated into the stock prices.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
