However, the company's own Twitter hedge fund was liquidated after just a month, and other similar ventures have also been short-lived. The fund made a 1.85pc return in its first and only month of trading. California-based MarketPsy Capital launched a fund trading on signals from interpreting social media and corporate speeches, but this was also closed.
Despite the closure of the DCM fund, which was mainly down to the tough trading environment, companies are racing to harness the information put out by social media – which typically gives a time advantage to financial traders. For example, Twitter put out the news of Osama bin Laden's death 20 minutes before traditional news outlets.
A social media start-up called Knowsis, launched last year, aims to give traders access to social media information. The company claims to use algorithms to amalgamate financially relevant social conversation into useful information for traders.
"Social media is an incredibly powerful platform and the finance world has taken notice. However, from a finance perspective it has its problems," said Oli Freeling-Wilkinson, Knowsis's founder. "It's unstructured, noisy and incredibly hard to keep track of the multiple streams out there. As a consequence, it is extremely difficult to extract value from it efficiently." He said this type of data would "only become more embedded as the next generation of traders and investors rise up through the ranks".
Other companies selling similar data include Dataminr, which provides similar information to companies and governments, and Sntmnt, a Dutch company that claims to reach accuracy levels of 62pc when predicting stock market movements using social media.
While these start-ups allow corporate entities to have access to social media sentiment, the DCM platform promises to offer it to ordinary investors. It promises to give a score of between one and 100 to a stock, portfolio or sector, based on social media sentiment.
"For the very first time we are connecting this information source to the trading community, opening up the universe of social media data so traders can make more informed buying and selling decisions," Mr Hawtin said. "The big benefit is that our sentiment rating adds another risk mitigation dimension to trading. For instance, if users see a downward movement in the sentiment rating of a stock in which they are perhaps slightly overweight, they can immediately hedge their position."
Academics have produced research that purports to prove that Twitter trends can predict stock market movements. Johan Bollen of Indiana University found that collective mood swings on Twitter could help predict movements in the US stock market.
He measured seven different types of mood and found that spikes in anxiety on Twitter were followed around three days later in dips in share prices. Mr Bollen then licensed the algorithm he used to DCM.
However, the company has now produced its own algorithm rather than using the licensed one.
No comments:
Post a Comment