Research Press Release

Big data: Can Google search volumes predict stock market movements?

Scientific Reports

April 25, 2013

A quantitative analysis of changes in Google query volumes for search terms related to finance, published in Scientific Reports this week, reveals patterns that may be interpreted as early-warning signs of stock market moves. The research may offer a new perspective on the behaviour of market participants during periods of large market movements.

Tobias Preis, Helen Susannah Moat and H. Eugene Stanley analysed Google searches from 2004 to 2011 to investigate the information-gathering process that precedes the trading decisions recorded in stock market data. The team focused on search volumes for a set of 98 search terms, including terms related to finance, debt and stock markets. Within the studied period, the analysis suggests that data capturing changes in Google search behaviour would have been able to anticipate certain future trends. The research supports the idea that drops in the financial market may be preceded by periods of investor concern. In these periods investors may search for more information about the market, before eventually deciding to buy or sell.

The study highlights the insights large behavioural data sets may provide into the processes of collective decision-making. Further work is needed to understand the underlying psychological mechanisms that may lead people to search for more information about financial topics before selling stocks at low prices.

DOI:10.1038/srep01684 | Original article

Research highlights

PrivacyMark System