Can Google Actually Predict Today's Stock Market?

Updated

A new study published in Scientific Reports says: "By analyzing changes in Google query volumes for search terms related to finance, we find patterns that may be interpreted as 'early warning signs' of stock market moves." Google Trends has been used to help track the spread of the flu and the rise and fall of fads, and now researchers claim it can be a useful tool for beating the stock market.

But plenty of questions remain that investors should ask before googling how to profit from the findings.

The study
Researchers took 98 common words, including "restaurant," "stocks," "religion," and "happy," and found a statistically significant relationship between the search popularity of many words and the weekly movement of the Dow Jones Industrial Average from 2004 to 2011.


The most telling word for future movement was "debt." The researchers formed a trading strategy based on this word, and while the Dow Jones returned just 16% over the seven-year period, the Google Trend strategy returned 326%.

Other supposedly predictive words and the standard deviations above a random-strategy return include:

Word

Cumulative Return (standard deviations)

debt

2.31

color

2.22

stocks

2.21

restaurant

1.7

portfolio

1.69

inflation

1.66

dow jones

1.56

revenue

1.51

economics

1.47

Source: Quantifying Trading Behavior in Financial Markets Using Google Trends, April 2013.

But before you take out your dictionary and start clicking, there are several things to think about.

The issues
While "debt" seemingly did a fantastic job at predicting the market and earning outsize returns, so did such unrelated words as "color," "restaurant," "religion," "cancer," and "marriage." One might hypothesize that optimistic traders search for restaurants more before a positive week of trading, but the other words point to data that has found relationships without a true connection. When testing so many words, some are bound to show some significance, even if there really is none. There are bound to be false positives.

The results seem coincidental: Even though "marriage" led to a relatively well-performing strategy, "ring" led to the worst-performing strategy.

Additionally, the strategy of buying or selling the market every week fails to take into account trading costs. Such a strategy would rack up not only brokerage fees, but also an investor's time, effort, and anxiety.

Stick with what you know
Even if this strategy has merit, this publication will likely have taken away any extra gains investors could tease out of these words. It's best to keep focusing on a company's fundamentals, value, management, and future prospects.

If you're looking for some long-term investing ideas, you're invited to check out The Motley Fool's brand-new special report "The 3 Dow Stocks Dividend Investors Need." It's absolutely free, so simply click here now and get your copy today.

The article Can Google Actually Predict Today's Stock Market? originally appeared on Fool.com.

Fool contributor Dan Newman has no position in any stocks mentioned. The Motley Fool has no position in any of the stocks mentioned. Try any of our Foolish newsletter services free for 30 days. We Fools may not all hold the same opinions, but we all believe that considering a diverse range of insights makes us better investors. The Motley Fool has a disclosure policy.

Copyright © 1995 - 2013 The Motley Fool, LLC. All rights reserved. The Motley Fool has a disclosure policy.

Advertisement