- News Home
17 April 2014 12:48 pm ,
Vol. 344 ,
Officials last week revealed that the U.S. contribution to ITER could cost $3.9 billion by 2034—roughly four times the...
An experimental hepatitis B drug that looked safe in animal trials tragically killed five of 15 patients in 1993. Now,...
Using the two high-quality genomes that exist for Neandertals and Denisovans, researchers find clues to gene activity...
A new report from the Intergovernmental Panel on Climate Change (IPCC) concludes that humanity has done little to slow...
Astronomers have discovered an Earth-sized planet in the habitable zone of a red dwarf—a star cooler than the sun—500...
Three years ago, Jennifer Francis of Rutgers University proposed that a warming Arctic was altering the behavior of the...
- 17 April 2014 12:48 pm , Vol. 344 , #6181
- About Us
Can Google Predict the Stock Market?
14 November 2010 7:01 pm
Whoever figures out how to predict the stock market will get rich quick. Unfortunately, the market's ups and downs ultimately depend on the choices of a massive number of people—and you don't know what they're thinking about before they decide to buy or sell a stock. Then again, maybe Google knows. A team of scientists has shown a strong correlation between queries submitted to the Internet search giant and the weekly fluctuations in stock trading. But it's unlikely to make anyone wealthy.
The stock market is a famously complex and jittery system. In any given week of trading, the price of shares in companies might stay the same, rise steadily, or suddenly crash. The causes of these patterns have evaded researchers, though not for lack of trying. An army of "quants"—many of them poached from academic math and physics departments—has studied data from stock indices such as the S&P 500 for decades. But within any given week, the time scale that matters to traders, the movement of the market seems random.
The reason, says Tobias Preis, a physicist at Johannes Gutenberg University Mainz in Germany, is that people decide to buy or sell stocks based not only on personal motivations but on the collective decisions of others. This "herding behavior" makes the stock market so chaotic that the pattern of trading in one week is nearly useless for predicting what will happen the following week. To predict the market, you need data on what is going through people's minds before they make their financial decisions. One such source of data is the total weekly volume of Internet search queries, now available to researchers through Google Trends.
Researchers led by Preis compared the week-by-week fluctuations in two sets of data: The number of times that the name of a company in the S&P 500 was included in a Google search query, and the price and trading volume of that company's stock. They focused on the 6 years from 2004 to 2010.
The findings, to be published 15 November in Philosophical Transactions of the Royal Society A, aren't going to make anybody rich. The Google data could not predict the weekly fluctuations in stock prices. However, the team found a strong correlation between Internet searches for a company's name and its trade volume, the total number of times the stock changed hands over a given week. So, for example, if lots of people were searching for computer manufacturer IBM one week, there would be a lot of trading of IBM stock the following week. But the Google data couldn't predict its price, which is determined by the ratio of shares that are bought and sold.
At least not yet. Neil Johnson, a physicist at the University of Miami in Florida, says that if researchers could drill down even farther into the Google Trends data—so that they could view changes in search terms on a daily or even an hourly basis—they might be able to predict a rise or fall in stock prices. They might even be able to forecast financial crises. It would be an opportunity for Google "to really collaborate with an academic group in a new area," he says. Then again, if the hourly stream of search queries really can predict stock price changes, Google might want to keep those data to itself.