Twitter chatter influences stock prices - survey

Twitter chatter influences stock prices - survey

Share price volatility is affected by the chatter of public opinion on social media sites such as Twitter and Facebook, according to UK financial professionals.

A poll of 360 people from Colt Technology Services found that 63% of respondents, including brokers and heads of trading desks, believe the valuation of individual stocks can be directly linked to public sentiment contained in social media channels.

Although only seven per cent regard sentiment on Twitter, Facebook and other sites as a leading indicator, just under half do consider it as a trailing indicator.

Some hedge funds and proprietary trading houses are trying to harness this social media chatter, feeding it into algorithmic strategies in a bid to steal a march on competitors who use more traditional forms of data.

The survey reveals some concerns over the use of sentiment-based trading, with nearly a third of respondents believing that the ability to respond fast enough to social media is a barrier to adoption.

The sheer volume of data from social media is seen as one of the major challenges to creating successful trading strategies, with 43% claiming that they would struggle to respond fast enough to the daily influx of information.

Regulatory pressures are seen as less of an issue, cited by just 13% as a barrier to adoption.

Hugh Cumberland, solution manager, payment and settlement services, Colt, says: "In a market where liquidity is highly valued and investors cautious, new sources of competitive advantage will always be welcome. What's important is working out how best to leverage the data mined from millions of social media messages to help trading firms cut through inertia and deliver much needed volume."

Earlier this month, one of the trailblazers of the social media sentiment trading world put itself up for auction after a brief, varied and mixed existence.

Derwent Capital launched a £25 million Twitter sentiment-based hedge fund in 2011 but the experiment was quickly shuttered despite a steady performance. In January it changed tack, offering the technology to day traders through a DCM Dealer platform.

Most recently it put itself up for auction on its Web site. A prospectus lists the guide price for the firm at £5 million, with a starting bid of £1. However, with less than five days to go until the auction closes, the highest bid is just £100,000.

Comments: (1)

Ketharaman Swaminathan
Ketharaman Swaminathan - GTM360 Marketing Solutions - Pune 15 February, 2013, 18:27Be the first to give this comment the thumbs up 0 likes

There are at least two challenges of mining messages from the entire universe of social media networks: (1) Mind-boggling volume (2) Lack of structure that makes accurate sentiment analysis all but impossible. That explains why there's renewed interest in tools that focus exclusively on Twitter messages and deliver adequate throughput at high accuracy levels.