Social Market Analytics Inc. (SMA) S-Factor metrics provide quantitative measures of the intentions of professional investors as expressed on Twitter. As a preferred partner of the social networking service, SMA monitors Twitter 24/7 to provide hourly sentiment indicators. These indicators are created using tweets from accounts that are deemed reputable by SMA's patented methodology.
S-Factor metrics offer raw sentiment and sentiment relative to a security’s history for all stocks on the Russell 3000, as well as for other major liquid securities listed on the NYSE, NASDAQ, and LSE.
Historical backtesting, independent research, and customer feedback have demonstrated that SMA offers a tradable, alpha-generating factor. The SMA S-Factor data feed represents a new uncorrelated source of predictive information that adds value to models.
Asset Class: Public Companies
Data Frequency: Event-driven
Delivery Frequency: Hourly
History: Data available back to 2011
Through its use of patented algorithms and security-specific topic models, SMA allows users to leverage valuable sentiment information from reputable Twitter accounts, while bypassing the noise that typically makes social media sentiment so challenging to aggregate. First, SMA analyzes the entire Twitter firehose using its patented financial NLP methodology and account rating algorithm. Topic models, which are a collection of inclusion- and exclusion-based rules customized for each security, are then used to determine if a tweet is about the given security.
SMA extracts all English tweets relating to a security based on its topic models, and then filters the tweets to those posted by reputable accounts that are identified by SMA’s 12-factor algorithm. Once this is done, SMA calculates sentiment for all eligible tweets and aggregates tweet scores per security to create S-Factors.
SMA’s S-Score metrics are used to compare the tone of the current conversation per security to historical levels. The S-Score is one of 15 primary S-Factors that is included in the product and used by SMA’s quantitative systemic trading clients.
Company-Level Sentiment Analysis
Track and analyze rapid changes in social media sentiment using SMA's S-Factors. These indicators supply consistent, quantifiable metrics that can be used to identify changes in a company's sentiment relative to its history. The metrics provide an early warning of breaking news and potential inflection points for a company.
An S-Score above 2 or below -2 indicates that the Twitter conversation is 98.6% more positive or negative over the past 24 hours compared to a 20-day baseline. Research has shown that when S-Score levels fall above 2 or below -2, securities start seeing movement in the direction of sentiment over the next one hour to three days.
The details provided above are as of April 2020.
If you have any questions or would like to learn more about any of the content mentioned above, please contact us at OFSupport@factset.com.