When will the stock market crash? As with most everything else, the answer lies within Google.
In a study published in the Proceedings of the National Academy of Sciences, researchers from Warwick Business School and Boston University have created a method to identify search terms that precede a stock market crash.
To help algorithms identify these patterns, the research team categorized every word on Wikipedia into different topics (e.g., words such as “business,” “management,” and “bank” fall under the business topic). The researchers then consulted Google Trends to find spikes in business- and politics-related queries between 2004 and 2012, and found a rise in certain terms can predict stock market falls.
“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,” Suzy Moat, assistant professor of behavioral science at Warwick Business School, said in a statement.
In particular, she said the link between political search terms and market activity “provides evidence that valuable information may be contained in search engine data for keywords with less obvious semantic connections to events of interest.” Beyond predicting stock market movement, the researchers say the development of this method can be used to create predictive models for other applications as well.AT