SMA Preps Social Media Sentiment Display for Asset Managers, Active Traders

Naperville, Il.-based Social Market Analytics, a provider of social media-based analytics and trading indicators, is preparing to release a real-time workstation of its social media sentiment scores, dubbed SmartStation (SMA Real-Time Station), which will provide traders and asset managers who want to incorporate the impact of social media into their trading strategies with a screening tool to access the vendor's real-time sentiment scores along with market price activity.

The vendor plans to initially release the workstation for a limited general audience next month, targeting portfolio managers who can upload portfolios and set alerts to monitor the impact of social media on individual securities, but who do not need a full datafeed of sentiment information, and active traders who want to use social media as an alpha-generating indicator. Pricing will be set according to how many sources a user follows, and how frequently they wish to receive updates.

"Right now, our product is a datafeed for professional investors. Quantitative analysts want all 3,000 securities every minute. But a portfolio manager might only follow 20 or 30 stocks and just want an early warning system. SmartStation will allow us to target a broader market and bring social media to people who have been left behind and who are getting killed because they have no way to incorporate social media," says Joe Gits, chief executive of SMA. "For professional investors, this is one of seven or eight inputs that they would use, while an active trader can use this to see where they will make money today."

Gits says the vendor's sentiment scores -- which include its S-Score weighted, normalized sentiment for a specific period; its S-Mean weighted average of sentiment scores for a given day; S-Delta percentage change in S-Score; S-Volume measure of indicative tweet volume; S-Buzz measure of unusual social media intensity, S-Volatility percentage measurement of sentiment variability; and S-Dispersion measure of tweet concentration -- are proven to provide a predictive indicator of positive or negative market reactions ahead of events such as earnings announcements, analyst downgrades or changes to a company's earnings outlook.

SMA sources social media data from Twitter's request API and from Boulder, Colo.-based social media aggregator Gnip's datafeed, then removes spam and duplicates -- though items are assigned a higher weight based on how many people are tweeting about them -- to focus on tweets from qualified professional investors that provide an indication of their intent. Currently this represents around 400,000 individuals, though Gits says indicative tweet volume -- based on the quality of tweets and number of validated accounts -- is growing by 10 percent every month.

Because upwards sentiment can persist for three days -- and around two days for short-side signals -- SMA's social media sentiment scores can alert traditional asset managers to developing situations affecting securities in their portfolios, in some cases 15 minutes earlier than regular news channels, or can provide another source of ideas for assessing whether to sell or hold a stock longer, Gits says.