By Andy Bloxham, The Telegraph,
Traders have used an awareness of the population’s collective emotions to predict price movements for years.
However, experts found that the instant nature of Twitter meant that those emotions could be gauged more accurately.
Previously, it had been thought that a drop in the markets led to more negative feelings but it proved to be the other way around.
Analysts at Derwent Capital Markets in Mayfair, central London, have launched a £25m fund that makes its investments by evaluating whether people are generally happy, sad, anxious or tired, because they believe it will predict whether the market will move up or down.
It takes a random 10% of all Twitter feeds and uses two methods to collate the data. One compares positive with negative comments and the other uses a program designed by Google, the internet giant, to define six moods calm, alert, sure, vital, kind and happy.
In a study published last October, Bollen used the social networking site to predict the direction of the movement of the Dow Jones in New York with 87.6% accuracy.
Mr Bollen’s algorithms flag up key emotive words when they appear in a certain order.
He told the Sunday Times: “We recorded the sentiment of the online community, but we couldn’t prove if it was correct. So we looked at the Dow Jones to see if there was a correlation. We believed that if the markets fell, then the mood of people on Twitter would fall.
“But we realised it was the other way round — that a drop in the mood or sentiment of the online community would precede a fall in the market.
“That was a eureka moment.
It meant we could predict the change in the market and that gives you a considerable edge.”
Paul Hawtin, Derwent’s founder and fund manager, has an exclusive contract with Bollen to use his technology.
Mr Hawtin told the newspaper: “Investors have always accepted that markets are driven by sentiment, mainly fear and greed. When people are greedy the markets go up and when they are fearful they go down.
“When sentiment dropped, and people tweeted about feeling tight on money, were worried or anxious, the markets would crash two or three days later.”
It is not the only such tool on the market.
WiseWindow, a data provider in California, boils down the online input of 100m social networking comments every month for its clients.