American and British researchers have developed an analysis method that reveals that a stock market decline is coming, and they believe that their technique can be used practically to provide warning signs before the event occurs in the real world of investment trading.
The "prediction" is based on an analysis of the content of searches on the Google search engine, given that, according to scientists, searches for business and political topics increase significantly before a "dip" in stocks.
The researchers created the appropriate software algorithms that can automatically correlate - without human intervention - the content of previous searches with the future movement of stocks.
Researchers from Boston University and the University of Warwick Business School, led by Chester Kerm, who published the relevant paper in the journal Proceedings of the National Academy of Sciences of the United States (PNAS), analyzed data for the period 2004-2012, correlating the performance of the shares of the American stock market index S & P 500 with the type of searches recorded by Google Trends.
"Search engines like Google record almost everything we search for. The records of these searches allow us to learn how people gather information online before they make decisions in the real world. So it's possible to use this search data to predict what large groups of people might do," Kerm said.
The key finding is that the more users around the world, investors and non-investors, search Google for a wide range of political, economic and business topics, the more they feel increasingly concerned about the future course of the markets. This growing lack of confidence eventually translates into reduced demand for stocks (or other financial products) and ultimately into a decline in their prices.
The researchers also pointed out that a similar analysis technique with the help of Google can be used to "predict" a wide range of other future events, beyond the course of the stock markets.
Source: tro-ma-ktiko.blogspot.gr
