Abstract:

With the advent of the Internet and the availability of user search query data on a broader scale, since the early 2000s researchers have started using collective search query information instead of, or, in addition to, traditional investor sentiment proxies. This study examines whether the leverage (bad news) effect, as measured by the EGARCH (1,1) model, changes with the inclusion of a newly emerging sentiment proxy, internet search volume. The sample consists of 14 US companies belonging to the NASDAQ and NYSE Indices and 501 observations of data collected at weekly frequency spanning a nine year period. Empirical findings suggest that, inclusion of the investor sentiment variable has no clear impact on the bad news effect; there is, however, a discernible increase in volatility persistence. The implications of the findings are that the investor sentiment proxy has additional informational content. Behavioral finance theory and the availability and social proof heuristics serve as potential explanations for such findings.

Keywords: