Using Twitter to predict the stock market

Tapping the zeitgeist on Twitter can be used to predict the stock market, new research has found.

Social networking platform Twitter tracks trends and moods across populations accurately enough to predict outcomes in purchasing, according to a study from Johan Bollen and Huina Mao from Indiana University in the US and Xiao-Jun Zeng from the University of Manchester in the UK.

Two mood-tracking tools were used to determine emotional sentiment. OpinionFinder measures positive versus negative emotions, while the other, Google-Profile of Mood States (GPOMS), allocates moods into six categories: Calm, alert, vital, kind and happy. The results were compared to Dow Jones data to determine if, and how, moods related to peaks and drops in the stock market.

Calm and happiness, isolated from the other variables, were found to significantly improve the accuracy of prediction models. “General happiness” measured by OpinionFinder was not found to have a predictive effect, the researchers say.

The study was inspired in part by studies showing public discussion in online forums can indicate book and movie sales.

[via Technology Review]

nextmedia Pty Ltd © 2022 All Rights Reserved