The Derwent Absolute Return Fund Ltd., established in starting operations in February with an initial 25 million pounds (39 million) under management, will the messages on the social networking website. A business model will highlight when the number of times the words on Twitter as "calm" to rise above or below average.
A document from the University of Manchester and the University of Michigan published in October said the number of emotional words on Twitter could be used to predict the daily movements in the Dow Jones Industrial Average. A change in the emotions expressed in line to be followed between two and six days later by a movement in the index, researchers, and this information to predict their movements with an accuracy of 87.6 percent.
"The mood and the mood changes dramatically the impact of positive and negative news," Hawtin said in a telephone interview. "If the market is in a very positive mood and optimistic, that the ability to ignore the bad news - bad news comes out and expects the Dow to fall, and it is not."
Twitter now has 175 million users and is 95,000,000 messages per day, according to its website. Has grown from 50 million a day from February, and researchers are finding new uses for this growing source of data in real time.
Trade patterns
Derwent Capital Markets, based in West End of London, signed an exclusive agreement this month with Zeng Xiao-Jun, a doctor of computer science at the University of Manchester, to develop research on business models. Derwent also meet with the associate professor and colleague Johan Bollen Huina Mao, University of Indiana this week to discuss the possibility of them working in the new fund, said Hawtin.
The new fund will use algorithms based on data from Twitter messages and other factors to trade the FTSE 100, FTSE 250 and the Dow Jones Industrial Average as well as oil, gold and other precious metals and coins.
Zeng, Bollen and Mao research measures the mood of the public by searching Twitter messages from February through December 2008 for synonyms and language related to six moods: calm, alert, safe, friendly and happy life . Then the researchers matched the time and date of these posts to the closing prices of the Dow Jones Industrial Average to test their hypothesis that changes in the sentiments expressed in line could predict future values of the index.
Calm Mood
Their results showed that rises and falls in the number of cases of words with a relaxed atmosphere can be used to predict similar movements in the price of closing, the Dow Jones of between two and six days later. The other states of mind was not as predictive quality, the paper said.
Derwent expected annual investment return of between 15 and 20 percent in the new fund, Hawtin, he said, adding that 1.5 million pounds of the initial bet is the money managers themselves.
"The only risk to us is if Twitter goes down and people do not use more of it," Hawtin said. "But we can only get bigger and better, and that more and more people are going to use to express their feelings."
Sixty-six percent of Americans online use social networking sites, according to a June 10 report by the web site s Experian Information Systems Inc. Facebook Inc. was the most visited in the world last week with 10.1 percent of all page views of the Internet, Experian, said on its website.
0 comments:
Post a Comment