People who embrace social media, particularly the wonderful world of Twitter, where users publish and receive ‘tweets’ of 140 characters or less, could be helping analysts to predict the rise and fall of share prices.

It comes with the news that millions of tweets posted by users of the Twitter social networking website are being harnessed by a British hedge fund to predict whether share prices will rise or fall.

The tweets, which range from what people had for breakfast and celebrity musings to companies making major announcements, are being analysed using a complex computer programme, to predict stock markemovements by assessing the public mood.

Analysts at Derwent Capital Markets in London have launched a £25m fund that makes its bets by evaluating whether people are on average happy, sad, anxious or tired, in the belief that this will predict whether the market will turn bullish or bearish.

The computer programme, developed in the United States, takes a random 10 per cent of all Twitter feeds and collects data in two ways. It compares positive with negative comments and, secondly, uses a program designed by internet giant Google, to define six moods – calm, alert, happy, sure, vital and kind.

In a study last year, Twitter was used to predict the daily swings of the Dow Jones share price index in New York with 87.6 per cent accuracy. The sentiment of the online community was recorded, in the belief that if the markets fell, then the mood of people on Twitter would fall.

However, researchers found that 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 the eureka moment that meant analysts could predict the change in the market and that could give them a considerable edge.

Now firms are branching out to analyse comments on blogs and other networking sites.

Outwrite PR