Every 20-something hedge fund employee is pitching their boss on arbitraging stock information on Twitter. A study in The Journal of Computational Sciences is giving those black box heads more fuel for the fire after revealing a “Twitter Hedge Fund” that used algorithms to track mood changes and trends in messages broadcasted via the social networking service then predicted an insanely high rate of accuracy in pegging stocks’ price changes across the day. The recently published findings state, “We find an accuracy of 87.6% in predicting the daily up and down changes in the closing values of the DJIA (NYSE:DIA) and a reduction of the Mean Average Percentage Error by more than 6%.”
Sounds like another case of momentum trading trumps financial analysis. Maybe the market isn’t as crooked and dominated by insider trading and hedge fund machinations as we’ve given it credit for.
You might have some new advice for your broker now … well, until the trade gets crowded and the edge is arbitraged away.