Tuesday, June 7, 2011

Twitter Predicts How Stocks Move 87.6% Of The Time


smiling_brokers

When a "Twitter hedge fund" was launched, we saw it as a bubble sign. But there seems to be something there. 

A Cornell University study in the Journal of Computational Sciences used algorithms to track the "mood" of messages on Twitter and found that certain moods on Twitter correlate very nicely with the closing of the Dow Jones Industrial Average.

"We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the DJIA and a reduction of the Mean Average Percentage Error by more than 6%," the study says. 

Don't get your hopes too up, however: as we know, correlation is not causation. While we're skeptical that "mood" on a global social network like Twitter can predict the movements of a specific index like the DJIA, it's certainly possible that some signals on Twitter can correlate with some stocks and other events. It's also the fastest way (probably even faster than a Bloomberg terminal) to track breaking news, which can have an impact on the stock market. 

Dare we say StockTwits has found its business model?

Don't Miss: NSFW: People Really "F---ing" Hate Twitter's New Email Notifications →

For the latest tech news, visit SAI: Silicon Alley Insider. Follow us on Twitter and Facebook.

Join the conversation about this story »

See Also:







Twitter Predicts How Stocks Move 87.6% Of The Time


Backlink: http://feedproxy.google.com/~r/typepad/alleyinsider/silicon_alley_insider/~3/EyDqrU6oxzg/twitter-predicts-how-stocks-move-876-of-the-time-2011-6

No comments:

Post a Comment