#Tweet this: To gain a trading edge, Wall Street traders are now using clever computer programs to monitor and decode the words, opinions, rants and e
ven keyboard-generated smiley faces posted on social-media sites like Twitter.com.
Human emotions, such as greed and fear, have always moved markets. Money can be made betting with or against the crowd, so measuring the mood of the masses online can be just as valuable as tracking price-to-earnings ratios, corporate profits and interest rates. The new trend on Wall Street for deciphering if the populace and investing public is in a positive or negative state of mind is computer-driven text analysis of the millions of real-time tweets and posts that flood social-networking sites.
Online surveillance of social-networking sites is emerging as a must-have tool for hedge funds, big banks, high-frequency traders and black-box investment firms that run money via computer programs. The goal: to gather market intelligence from previously untapped sources.
This emerging tool works something like this: Are you feeling glum, fearful or anxious today? Or are you in a calm, happy, optimistic mood? If you share your state of mind with the digital world via a tweet on Twitter or a soul-baring post on Facebook, Wall Street probably knows how you — and millions of other people — are feeling, too, thanks to its growing use of linguistic analysis of online posts. This incoming psychological snapshot of the Twitterati, digerati and average Joe could prompt a computer program interpreting the data at a hedge fund to place a trade without human intervention in an attempt to profit from the information.
The conversation tying tweets and social media and stocks together has gotten louder ever since last fall, when an academic study at Indiana University in Bloomington found a correlation between the collective mood of millions of people identified by tweets and the direction of the Dow Jones industrial average.
How the Twitter trade works
1. Capture daily tweets.
2. Analyze tweets with mood-measurement tools.
3. Determine mood content of tweets (i.e., “positive” vs. “negative” or “calm” vs. “anxious”).
4. If mood is positive or “calm” readings jump, buy stocks, because Dow is likely to rise three to four days later.
5. If mood is negative or “calm” readings plummet, sell stocks, because Dow is likely to fall three to four days later.
(IU study showed 86.7% accuracy rate in predicting the direction of the Dow three to four days later.)
Source: USA TODAY research
“We are in the early stages of a gold rush,” says Johan Bollen, a professor of informatics at Indiana University and co-author of the study linking Twitter mood measurement to stock market performance. “If you would have told anyone 10 years ago that this data would be available, they would have called it science-fiction. We know that emotions play a significant role in markets,” he says. Analyzing millions of tweets is akin to a “large-scale emotional thermometer for society as a whole.”
Derwent Capital Markets, a London-based hedge fund, was so taken with Bollen’s findings that it will soon launch a fund based on the methodology in his paper.
Interest in the marriage of social media and finance remains high. In March, a study done by a Ph.D. candidate at Pace University showed a positive correlation between stock price performance and the social-media “popularity” of well-known brands Starbucks, Coca-Cola and Nike. The Pace author, Arthur O’Connor, also found that brand popularity online may be a “lead indicator” of stock performance. And a team of economists at TUM School of Management, or Technical University of Munich, has created a website, TweetTrader.net, that attempts to profit from similar Twitter research.
For more info: http://usat.ly/irbg0U