Research suggests Google can predict the stock market

Findings by Warwick University academics has revealed that an analysis of volume changes in Google search terms related to finance could be early indicators of stock market movements. 

A team of academics looked at changes in the Google search frequency of 98 terms, including ‘revenue’, ‘unemployment’, ‘credit’ and ‘nasdaq’, from 2004 to 2011.

Their research suggested that a heightened interest in financial topics implied increased investor concern, and may be followed by drops in the financial market. Investors may search for more information about the market before selling at lower prices.

Respectively, drops of interest in financial topics could be used as a signal for subsequent stock market rises.

They found that using changes in search volume as a trading strategy to invest in the Dow Jones Industrial Average Index could lead to substantial profit.

In their report entitled “Quantifying Trading Behavior in Financial Markets“, it was demonstrated that trading on the basis of the Google search volume of the term ‘debt’ could bring profit returns of up to 326 percent compared to the conventional Buy and Hold Strategy.

The findings were conducted by Tobias Preis, associated professor of Behavioural Science at Warwick Business School, Helen Moat of University College London, and H. Eugene Stanley of Boston University.

Dr Moat said: “Analysis of Google Trends data may offer a new perspective on the decision making processes of market participants during periods of large market movements.

“It’s exciting to see that online search data may give us new insight into how humans gather information before making decisions – a process which was previously very difficult to measure.”

Dr Preis added: “We are generating gigantic amounts of data through our everyday interactions with technology. This is opening up fascinating new possibilities for a new interdisciplinary ‘computational social science’.”

The study was part of the IARPA Open Source Indicators programme, which aims to develop methods to predict significant events through continuous, automated analysis of publicly available data.

Jason Matheny, manager of the programme, said: “This work illustrates the insight that publicly available data can provide to identify early warning signals of emerging events in the world.”