A recent Money Week article studies whether web based data from search engines and social media can be used to predict future behaviour. Similarly, other academic researchers have discovered that mentions of political candidates on Twitter are good predictors of electoral outcomes.
The article refers to a paper entitled Twitter Mood Predicts the Stock Market published last year whereby three computer scientists researched whether societies experience collective mood states that affect the nation’s decision making and whether these correlate and can predict economic indicators. By measuring the ever growing stream of words that emanate from the Twitterverse, the study found that rises and falls in instances of certain words indicative of mood correlate with rises and falls in the stockmarket between two and six days later which seems to confirm the long held belief that stock prices are influenced by the general public feeling.
As a result of this study and working with its authors, London based hedge fund Derwent Capital Markets has based an entire fund (the Twitter Fund) on using social media analytics to guide their thinking. The fund, the first of its kind, was launched in May 2011 and bases its decisions principally on “sentiment derived from real-time social media data analysis” in an attempt to use technology and data to be able to quantify human emotion. So far the fund has recorded positive returns for investors and outperformed the wider hedge fund industry. Paul Hawtin, Derwent’s founder and fund manager, noted that when sentiment dropped, and people tweeted about feeling tight on money, were worried or anxious, the markets would crash two or three days later.
There are a growing number of other studies looking at social media as a lead indicator of share price performance which indicates that brands will allocate more and more resources to ensure that there is a positive online buzz surrounding their name in order to improve their stock performance as traders start to follow the online metrics ever more closely.