July 30 ,2014 | by Hari Srinivasan

Could Google search predict a market crash?

Google search predict a market crash

Popular Google search queries could potentially be used to predict falling stocks, new research has found.

Every analyst and trader on Wall Street wants to stay ahead of the game and avoid the next great crash. But researchers at Warwick Business School and Boston University say they have developed a method that could potentially use one of the world’s biggest data sources – Google.


The study published in the Proceedings of the National Academy of Sciences identified that a rise in the number of searches for business and politics topics came immediately before falls in the stock market.

According to the research team, the method could potentially help to detect the warnings signs before a range of different major world events using search data alone.

“Records of these search queries allow us to learn about how people gather information online before making decisions in the real world,” says Chester Curme, research fellow at Warwick Business School and lead author on the study.

“So there’s potential to use these search data to anticipate what large groups of people may do.”

He added that there is such a wide range of potential search queries that the real challenge lies in working out which searches will be relevant to different events.

The study itself was huge, with the team having to categorise every word on Wikipedia according to its meaning.


Using Google Trends data they determined how many specific words were searched in the US from 2004 to 2012 and applied that data to a trading strategy for the S&P 500 index.

Changes in the search frequency for words related to business and politics were often followed with movements in the market.

Suzy Moat, assistant professor of behavioural science at Warwick Business School, explained that the results back up the hypothesis that business and politics searches are linked to uncertainty about the economy, which then translates into the selling of stock.

But Ms Moat’s colleague Tobias Preis, also an assistant professor of behavioural science at Warwick Business School, cautions that the link has weakened with time over the course of the study, so it may be that more complex strategies are required to use this information in trading.

Hari Srinivasan

Hari is the LSBF Blog's News Editor. He manages the editorial content on the blog and writes about current affairs, SME, entrepreneurship, energy, education and emerging market news.

Share on Facebook Share on LinkedIn
There are no comments posted yet. Be the first one!
Please write your comment, minimum length 50 characters
Please insert your name
Please insert a correct email address
We couldn't process your comment, please try again later