Policy Research

Quantifying Trading Behavior in Financial Markets Using Google Trends

Tobias Preis, Helen Susannah Moat, H. Eugene Stanley

Summary

The authors used google trend data, which “provides access to aggregated information on the volume of queries for different search terms and how these volumes change over time” to examine the relationship between this big data source, and the stock market. The authors used a complex statistical model to examine the relationship between particular search terms, and stock market movements. They found that google trends not only accurately predicted the current state of the market, but also provided insight into future stock market movement as well. Essentially, the authors selected certain financial terms, such as ‘debt’, and examined the search volume relative to particular trends in the market. They found that these search terms were more highly searched for preceding a market movement. Further, the authors used these search terms, and their predicted impact of the movements on the market to create a successful future trading strategy that would have been quite profitable if implemented.

Read Now

Policy Implications

The policy implications for this particular work are more complex than with other big data sources. First, legislators need to understand the usefulness of big data, and its implications in the U.S. and the world. As more and more big data sources become available, certain economic aspects need to be considered. For example, it would be possible to manipulate big data trends to manipulate the stock market, and as such, legislators need to become data savvy to access these risks. Further, in this particular case, not only do legislators need to protect against manipulation of big data, but also the usefulness of such data. For example, if legislators had a warning regarding the 2008 financial crisis, which this study indicates was possible with this data, the crisis may have been completely different.