Want to check the markets? See what's trending...
Support truly
independent journalism
Our mission is to deliver unbiased, fact-based reporting that holds power to account and exposes the truth.
Whether $5 or $50, every contribution counts.
Support us to deliver journalism without an agenda.
Louise Thomas
Editor
For a six-year-old, Twitter has achieved some impressive things. The social-networking site, which celebrated its sixth birthday last week, has already played its part in the Arab Spring and helped to stir up political scandals. Now some believe it could aid investors playing the stock market.
Researchers at the University of California have found a direct correlation between the number of conversations taking place about a company on Twitter and the volume of trading on its stock. Using the data to simulate a Twitter-based trading strategy, the team found that their model was better at predicting stock-price fluctuations than established market indicators.
The crucial variable was the number of what researchers called "connected components" – separate conversations on distinct topics relating to a company. For instance, conversations about the new iPad, Steve Jobs and iTunes would all be connected components of Apple.
Stock price was also found to be slightly correlated with the number of connected components. Basing trading strategy on their findings, a team led by Vagelis Hristidis, associate professor of computer science and engineering, created a model that outperformed the Dow Jones Industrial Average in a simulated four-month trial, with a 2.2 per cent loss for the Twitter model, against a 4.4 per cent loss for the Dow.
The study is not the first to look into the relationship between the stock market and Twitter, which now has 140 million users making 340 million posts a day. In 2010, sociologists at Indiana University Bloomington claimed that the mood of Twitter users on a given day could predict stock market changes three days later with 86 per cent accuracy. A number of hedge funds now analyse social-media trends to predict the market. However, there has been limited research into the impact on the stock market.
Professor Hristidis theorised that the volume of trading correlated to the variety of conversations taking place about a company, because discussions about a piece of bad news on Twitter would all follow a specific thread, whereas good news would attract less-focused attention and conversations would diversify into wider discussions about the company.
Subscribe to Independent Premium to bookmark this article
Want to bookmark your favourite articles and stories to read or reference later? Start your Independent Premium subscription today.
Join our commenting forum
Join thought-provoking conversations, follow other Independent readers and see their replies
Comments