The different impacts of news-driven and self-initiated search volume on stock prices

Big data benefits both Internet finance and behavioral finance research; Internet search frequency on stocks has been widely used to measure investor attention. In this study, we divide the search volume into news-driven and self-initiated by the online media coverage collected from Baidu Index. In a sample of CSI 300 stocks from 2009 to 2013, we find that self-initiated (news-driven) search volume is more likely to generate buy (sell) pressure, and media coverage can negatively moderate the impact of search volume on stock prices, suggesting that distinguishing search environment for investors can help improve the measure for investor attention.