A recent startup called SNTMNT has opened its Trading Indicator API, which provides price predictions about the price of S&P 500 stocks. The company says its accuracy is about 64%, which suggests that users of the API will be consolidating these indicators with other sources of information.
The approach is a very detailed level of analysis of Twitter sentiment about the stocks, based on so-called ‘stock tickers’ — references to companies like AT&T, whose ticker is T. Many Twitter users talk about stocks, referring to the companies by ticker, which is preceded in most cases by a dollar sign, as in $T.
Johan Bollan and other researchers published a paper, Twitter mood predicts the stock market, back in 2010, which was widely discussed. Their work suggests that the general mood on Twitter predicts the rise or fall of the daily closing price of the Dow Jones Industrial Average with 86.7% accuracy. SNTMNT is strongly influenced by that work, and is another indicator of the growing possibilities in computational social science.
My prediction is that sentiment analysis at either the macro- or micro-level — predicting stock market aggregate moves or the trends for specific stocks — will become a commonplace over the next few years, one of the most obvious applications of big social data.
Now, if someone could only predict the weather…
3 Comments
As we all know Sentiment mining has a ways to go, http://www.ignitesocialmedia.com/social-media-monitoring/radian6-sentiment-analysis-review/ but companies like lexalytics http://www.lexalytics.com/, idillia http://www.idilia.com/ are making great strides.
What I appreciate here is that they aren’t exaggerating their accuracy measures… 64% is about what we measured for some of the leading providers.
In the future as we are moving towards the semantic web entity extraction and text disambiguation technologies will be used before sentiment mining and we should see a big improvement in accuracy of sentiment.
How long will it be until someone tries to pump and dump via Twitter? Or use a few thousand bots to short and distort?
While most of the research and hype to date has been about stocks, there may be more value in mining for early economic indicators that drive financial performance.
Similarly, weather is impossible to predict 1-2 weeks out, but long-term climate change is much better understood.
Hi Daniel, I think pump and dump has been happening on Twitter with traditional IR firms since Twitter became important to PR firms, back in 2009-2010.
A Toronto company, that I believe was cleared, was investigated by the OSC a few years back:
http://www.theglobeandmail.com/globe-investor/agoracoms-online-chatter-case-goes-before-regulator/article1241685/
Looking into crunching the big data around social in general combined with a robust data cleaning process and new algorithms could be a sound way to improve existing financial indicators, or more probably, create new ones.