How Twitter Can Improve The Efficiency Of Markets

The content published on Twitter has enabled researchers to do a great many things, not least of which is to accurately predict trends as they unfold.  New research, led by the University of Toronto, suggests that Twitter could prove equally valuable in improving the efficiency of financial markets.

The authors believe that the collective, ‘hive mind’ of Twitter can do a very effective job of predicting company financial performance, and subsequently their stock market share price.

“If you want to find out if a stock or company is likely doing well or doing badly, you don’t just follow what the person in The Wall Street Journal or Barron’s magazine is saying,” the researchers say, “you also focus on what other people who are also investing in stock markets are saying.”

Market sentiment

The researchers performed textual analysis to around 870,000 unique tweets that talked about 3,600 different companies between January 2009 and September 2012.  The analysis revealed that Twitter activity leading up to a company’s quarterly earnings announcements accurately predicted whether the company would meet their targets or not.  What’s more, this activity was also good at predicting the way the company’s share price would respond.

Twitter was especially good at predicting such movements when there was little information about the company, which the researchers believe highlights the valuable role social media can play in promoting market efficiency by ensuring valuable information is shared.

“The tweets are from both people expressing their own original opinions about firm fundamentals and firms passing on existing information,” the researchers say. “This highlights the dual role of Twitter—as a source of new information and as a new mechanism to disseminate existing information.”

Suffice to say, there has been considerable concern about the potential for misinformation to spread on social media, and especially for rogue actors to deploy Twitter bots to help create and spread that misinformation.  The researchers believe that their findings suggest that such concerns are perhaps over-emphasized, and that the wisdom of crowds effectively filters these rogue voices out.

Whether that is sufficient to give people confidence in using Twitter as a market prediction tool however is still up for debate.  Nonetheless, the findings are certainly interesting.

Related

Facebooktwitterredditpinterestlinkedinmail