Predicting the Stock Market with Google Search Terms
'Big Data' researchers have demonstrated the possibility of creating money-making investment strategies for the stock market by mining Google search terms related to finance. They claim that their strategies would have yielded overall profits several times higher than a conservative strategy of buy-and-hold.
The research team led by Professor Tobias Preis of Warwick Business School in the U.K., Helen Susannah Moat of University College London, and H. Eugene Stanley of Boston University, have christened their investment strategy, the “Google Trends Strategy.” These big data academics analyzed a broad range of 98 commonly used search terms that range from “unemployment” to seemingly non-related words “water” and "marriage", and simulated investing strategies based on week-by-week changes in the frequencies of each of these words as search terms by American Internet users.
The study was conducted over a period of eight years, from 2004 to 2011. The team claimed that by using the word "debt" and the subsequent derived investment strategy, they would have generated a profit of 326 per cent over the research period, when compared to a profit of just 16 per cent if the buy-and-hold investment strategy was used instead. The work of Professor Preis and his team were funded by the US government funded Open Source Indicators program, which is housed within the Intelligence Advanced Research Projects Activity (IARPA).
(Source: Bloomberg Businessweek)