“…Typically, investor sentiment on social media is extracted from user posts and aggregated at the asset level (Das and Chen, 2007), and it has been widely used to demonstrate its predictability of financial market performance, such as the trends of Dow Jones or S&P 500 Index (Bollen et al, 2011;Zheludev et al, 2014;Piñeiro-Chousa et al, 2016), stock price movement (Oh and Sheng, 2011;Zhang et al, 2012;Wang et al, 2015), abnormal returns (Ranco et al, 2015;Deng et al, 2018), earning surprises (Chen et al, 2014;Bartov et al, 2018), trading volume (Tan and Tas, 2021), and market volatility (Hou and Tripathi, 2015;Audrino et al, 2020). However, mixed evidence is also reported that investor sentiment does not have strong predictability of stock returns (Oliveira et al, 2013;Kim and Kim, 2014), or the magnitude of the effect is economically small (Nofer and Hinz, 2014).…”