Can Google Tell Us When the Market Will Crash Next?
We all knew it was only a matter of time before people learned how to use the Internet to predict the future. The analysis of data can lead to insightful predictions, and thanks to companies like Google (NASDAQ:GOOG), we’re drowning in data.
Google Search in particular is a treasure trove of information for would-be soothsayers. Someone’s search history can reveal a tremendous amount of information, much of it accurate. Perhaps the most commercialized application of this is in targeted advertising, which is not only a cornerstone of Google’s business, but a practice championed by other massive data-producing engines like Facebook (NASDAQ:FB) and Amazon (NASDAQ:AMZN).
Not to get too Orwellian, but nearly every action and decision made by somebody online is recorded and crunched. This is the mark of the information age, and people have taken the torch and run with it. Every day there is a new application for big-data analytics. Investors are no stranger to these applications, but to date, most attempts to use big-data analytics to try to forecast financial market trends have either failed gloriously (remember the hedge fund that was set up to trade based on information about stocks and markets curated from Twitter?) or have taken root in the murky world of high-frequency trading.
This all could change, though. A study performed by a group of U.S. and U.K. academics has found that by studying searches of financial terms on Google, checked against the frequency of a set of financial terms in the Financial Times, could yield a winning strategy.
The study correlated the search volume for 98 financial terms to the closing price of the Dow Jones Industrial Average. The team presented their findings using “debt” as an example. Between 2004 and 2011, the study showed that a trading strategy built around increases or decreases in search volume for the word would have returned 326 percent, versus just 16 percent for the wider market.
At a glance, the correlation is clear: The more people are worried about their financial situation, the more researching they are going to do. This indicates that Mr. Market may be feeling nervous, and that investors are in a selling mood. As search volume went up, markets generally moved down — and the opposite is true, as well. Investors seem less interested in the markets if they are doing well.
This is certainly an interesting correlation, and the Financial Times reports that the researchers are in talks with investors over any possible real-world application of the technique.
It’s important — and also interesting — to point out that the study didn’t necessarily crack a secret code. Other seemingly random correlations also cropped up during the process. For example, a trading strategy building around searches for “colour” or “restaurant” appeared to be better indicators of markets than “debt,” which was the best-performing financial term.
Don’t Miss: Are Macs Better Than PCs at Being PCs?