A new study made waves last week when it announced that it could, in a sense, predict which way the Dow Jones Industrials (DJINDICES:^DJI)would go by analyzing Google (NASDAQ:GOOGL)search terms. The Warwick Business School in England and Boston University's department of physics' study, published in Nature's Scientific Reports, showed that investing in a hypothetical portfolio by analyzing terms in Google Trends from 2004 to 2011 yielded the researchers a 326% return.
But before you open up a new browser tab and call up Google Trends, let's talk about how this might not be the best investing strategy.
True science, false expectations
It's not wide to dismiss the merits of science, but it's smart to put the study's findings in context. Researchers used 98 search terms over the course of seven years to determine whether the Dow would go up or down after certain search terms were used.
Let's think about this for a moment. They studied 98 different words varying from "happy" to "politics" over the course of seven years to find how they may affect the Dow -- that's a pretty large amount of data for any investor to consider. Gathering all that data, sifting through it, and then trying to make investment decisions off it would make reading an annual 10-K seem like a walk in the park.
Over the course of the study, the research team found that the term "debt" was the best predictor of what the Dow would do. Increased searches for the term led to a selling off in stocks the following week, while a decrease in the search term indicated that the Dow would rise the next week. While this is a very interesting correlation, deciding when to invest your hard-earned money based on one search term isn't advisable.
Using a given term to predict market fluctuations today can't always be used to determine future predictions. The overall perspective of a given population can change, and its worries and enthusiasm can shift. Searches for "debt" may no longer be relevant, while searches for "zombie apocalypse" may indicate serious market fluctuations in the near future. At least that's what this writer is betting on.
Aside from combing through all of the data and finding an applicable search term to make investing decisions from, the problem of collective outside influence comes into play. Simply put, what people search for and how that translates into what's happening isn't always going to be true. For example, Google's Flu Trends is sometimes used to predict how much of the U.S. population is infected. But while Google's site said 11% of people had the flu back in January, the Centers for Disease Control said only about 6% actually had the flu. That's a pretty significant difference, and one that speaks to the level of variation between online data and offline facts.
A better approach
The stock market study is an important one to consider in the context of big data and how we use information to make decisions. But investing based on its findings is ill advised at best. It's easy to throw money into hypothetical portfolios, but much more difficult to lose real money when your search term didn't pan out.
The Motley Fool's investing perspective has always been about taking a long-term approach to investing in great companies. We don't all agree on what those companies are, but by doing our due diligence, listening to Motley opinions, and yes even some Google searches, we believe there are great investments to be had. At least until the zombies take over.