Google searches can predict stock market crashes - study

Google searches for various business and politics topics can predict future stock market crashes, according to a new study.

Researchers at the Warwick Business School analyzed search terms between 2004 and 2012, finding an increase in internet searches preceded market falls.

The group used a complex computer program to mine through a vast amount of internet traffic to spot trends attached to a multitude of topics or sectors.

“Search engines, such as Google, record almost everything we search for,” said Research Fellow Chester Curme, according to the Daily Mail. “Records of these search queries allow us to learn about how people gather information online before making decisions in the real world.”

Curme said this kind of analysis could be used to predict major stock crashes before they happen.

“So there’s potential to use these search data to anticipate what large groups of people may do,” he said. “However, the number of possible things people could search for is huge. So an important challenge is to identify what types of words may be relevant to behaviours of interest.”

The team quantified every word on Wikipedia to enable algorithms to automatically sense patterns in searches that could be related to real world behavior.

Thus, researchers could categorize words into topics so that a “business” topic could contain words such as “business,” “management,” and “bank.”

The team then used Google Trends to monitor how often these thousands of words were searched online in the United States between 2004 and 2012.

They found that changes in how often the politics and business terms were searched for could be connected to stock market activity.

“By mining these datasets, we were able to identify a historic link between rises in searches for terms for both business and politics, and a subsequent fall in stock market prices,” said Assistant Professor of Behavioural Science Suzy Moat. “No other topic was linked to returns that were significantly higher than those generated by randomly buying and selling.”

Moat said their method offers a new approach to identifying “less obvious” keywords.

“The finding that political terms were of use in our trading strategies, as well as more obvious financial terms, provides evidence that valuable information may be contained in search engine data for keywords with less obvious semantic connections to events of interest.”

Moat added that increased interest in financial search terms translated to particular stock market activity.

"Our results are in line with the hypothesis that increases in searches relating to both politics and business could be a sign of concern about the state of the economy, which may lead to decreased confidence in the value of stocks, resulting in transactions at lower prices."

Researchers from Boston University were involved in the study, which was first published in the Proceedings of the National Academy of Sciences.