What’s the old adage? Buy low, sell when Twitter users are in bad moods? Is that it? If not, maybe it should be, because according to a story in Wired, "[t]he emotional roller coaster captured on Twitter can predict the ups and downs of the stock market, a new study finds. Measuring how calm the Twitterverse is on a given day can foretell the direction of changes to the Dow Jones Industrial Average three days later with an accuracy of 86.7 percent.
The findings were somewhat stumbled upon, according to Johan Bollen, the social scientist behind the study. Attempting to find the mood of the public through Tweets, Bollen and Huina Mao, a grad student, used a questionnaire aimed to attach feelings to adjectives. After searching millions of Tweets for those adjectives—and other words used in conjunction with them—Bollen and Mao figured they could see the general mood of the population—at least those on Twitter.
Using this information along with an algorithm trained to predict the fluctuations of the stock market, the algorithm’s accuracy increased to 86.7 percent from 73.3 percent. That is, when the information of the general mood on Twitter was taken into account, this algorithm was able to predict much more accurately which way—up or down—the stock market would go.
Bollen admits that more research is needed to understand why this happens, but until then, why not add Twitter to your list of resources for figuring out who’s best to play with your money?
Image by mil8, licensed under Creative Commons.