Last week there was a report that came out from Indiana University (see here), where researchers had come up with a method for predicting changes in the Dow Jones through an analysis of Twitter updates. The researchers analyzed 9.7 million tweets from March to December 2008 with the use of two mood-recording algorithms, the Google-Profile of Mood States (GPOMS) and OpinionFinder.
What they found was that there were strong correlations between the moods and the movement in the market. Supposedly, there was a correlation between the "calmness" index (one of the six moods that was measured in GPOMS) and this could be used to predict whether the Dow Jones went up or down between 2-6 days later.
"We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the DJIA and a reduction of the Mean Average Percentage Error by more than 6%. "
Incredible? I think so too. Articles that I've read said that they need to do more research on this; however, the impact of social media is undeniable.
This is brilliant, I would like to know more about this. This could be either really really good, or very bad for the stock market industry.
ReplyDelete