Widespread Worry and the Stock Market - CiteSeerX

Widespread Worry and the Stock Market Eric Gilbert and Karrie Karahalios Department of Computer Science University of Illinois at Urbana-Champaign [egilber2, kkarahal]@cs.uiuc.edu Abstract risk-averse. Still, this thread of research comes from the Our emotional state influences our choices. Research on lab. How do real world emotions affect real world markets, how it happens usually comes from the lab. We know rela- like the stock market? tively little about how real world emotions affect real world In this paper, we take a step toward answering this ques- settings, like financial markets. Here, we demonstrate that tion. From a dataset of over 20 million LiveJournal posts, estimating emotions from weblogs provides novel informa- we construct a metric of anxiety, worry and fear

called the tion about future stock market prices. That is, it provides information not already apparent from market data. Specifi- Anxiety Index. The Anxiety Index is built on the judge- cally, we estimate anxiety, worry and fear from a dataset of ments of two linguistic classifiers trained on a LiveJournal over 20 million posts made on the site LiveJournal. Using a mood corpus from 2004. The major finding of this paper is Granger-causal framework, we find that increases in expres- that the Anxiety Index has information about future stock sions of anxiety, evidenced by computationally-identified market prices not already apparent from market data. We linguistic features, predict downward pressure on the S&P demonstrate this result using an econometric technique 500 index. We also present a confirmation of this result via called Granger causality. In particular, we show that the Monte Carlo simulation. The findings show how the mood Anxiety Index has novel information about the S&P 500 of millions in a large online community, even one that pri- index over 174 trading days in 2008, roughly 70% of the marily discusses daily life, can anticipate changes in a seem- trading year. We estimate that a one standard deviation rise ingly unrelated system. Beyond this, the results suggest new ways to gauge public opinion and predict its impact. in the Anxiety Index corresponds to S&P 500 returns 0.4% lower than otherwise expected. This finding is not as farfetched as it may first appear. In Introduction a 2007 paper, using Granger-causal methods, Paul Tetlock demonstrated that pessimism expressed in a high-profile Fear is an automatic response in all of us to threats to our Wall Street Journal column had novel information about deepest of all inbred propensities, our will to live. It is also Dow returns from 1984 to 1987. Nice, sunny weather even the basis of many of our economic responses, the risk aver- explains some stock market movements (Hirshleifer and sion that limits our willingness to invest and to trade, espe- Shumway 2003). Google search queries have predictive cially far from home, and that, in the extreme, induces us to disengage from markets, precipitating a severe falloff of information about diseases and consumer spending (Choi economic activity. (Greenspan 2007, p. 17) and Varian 2009). Blog posts and blog sentiment have been shown to predict product sales (Gruhl, et al. 2005; Mishne The stock market usually reflects business fundamentals, and Glance...

Website: comp.social.gatech.edu | Filesize: 478kb
No of Page(s): 8
Download Widespread Worry and the Stock Market - CiteSeerX.pdf