Can Facebook Predict Stock Market Activity?

Abstract

Using a novel and direct measure of investor sentiment, I find that Facebook's Gross National Happiness (GNH) has the ability to predict changes in both daily returns and trading volume in the US stock market. For instance, an increase of one standard deviation in GNH is associated with an increase of 11.23 basis points in market returns over the next day. Consistent with noise trader models, the influence of GNH on market returns is temporary and is reversed during the following trading weeks. I also verify the empirical validity of GNH by performing several tests in different natural settings.

Keywords: Investor sentiment, social media, behavioral nance, Facebook, social networks

JEL Classification: D81, G11, G12