How Researchers are Using Twitter to Predict Stock Market Changes

How Researchers are Using Twitter to Predict Stock Market Changes

A wild headline courtesy of researchers at Indiana University, who are using mood-recording algorithms on Twitter to predict stock market changes!  From MIT’s Technology Review:

So these guys took 9.7 million tweets posted by 2.7 million tweeters between March and December 2008 and looked for correlations between the GPOMS indices and whether Dow Jones Industrial Average rose of fell each day.

Their extraordinary conclusion is that there really is a correlation between the Dow Jones Industrial Average and one of the GPOMS indices–calmness.

In fact, the calmness index appears to be a good predictor of whether the Dow Jones Industrial Average goes up or down between 2 and 6 days later. “We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the Dow Jones Industrial Average,” say Bollen and co

That’s an incredible result–that a Twitter mood can predict the stock market–but the figures appear to point that way.

stock market twitter predicting changes

Fascinating story, and it’ll be interesting to see this explored in more rigorous detail.

Is this a case of life imitating art?  From former hedge fund manager Andy Kessler’s recent book, Grumby:

In one marathon coding session – I turned up the oxygen content in the middle of the night – I bang out arBroker. I am careful to not mix the mood code that determines individual risk profiles and the code that goes out and sums aggregate expectations of all our users about products and companies. I had to be really careful about whose mood I was interested in before issuing a ‘buy’ or a ‘sell’. Markets deal in aggregates but at the end of the day, individuals make all the decisions. I saw that on the floor of the stock exchange, at Fidelity’s offices and in my crude understanding about markets. Millions or billions of decisions are summed and calculated to determine the right price at that particular moment in time.

I can’t wait to fire it up, and fortunately, the market opens at 6:30 a.m. on the west coast, so I run out to Starbucks and buy a Venti Mocha Double Latte Extra Hot No Whip and I’m good to go and test this sucker out.

Thanks to Meeta, our database is filled with all sorts of formerly useless trivia, on buying habits of Grumby users, to questions asked at stores to real time reviews of products. It’s not really hard to find this stuff in the haystack of Grumby-collected data and aggregate it to match the mood of an individual.

As I load up arBroker, I take a long sip of my latte and then let out a long exhale of delight.

You just bought 300 shares of Starbucks Corporation, symbol SBUX at twenty five dollars and forty two cents,” my Grumby informs me.

Whoa. That was fast.

You can listen to our recent interview with Kessler here.

We’ve explored social mood and stock prices before – for some extra fun reading on the topic, please checkout: