Social media help predict stock prices. This is evident from research by Laurens van Leeuwen, a student at the University of Amsterdam, who investigated the predictive power of social media for the Dutch stock exchange AEX for his thesis. The research is a logical follow-up to previous studies, which showed how social media can be used to predict, among other things, film revenues, book sales and even national elections. By looking at how AEX funds are discussed on various social media platforms, 35% of all stock market fluctuations (up and down) can be predicted 4 days in advance. The more a particular fund is discussed, the greater its predictive power. The trading volume also increases as a fund is discussed more often.
Sentiment analysis
For the research, Laurens van Leeuwen uses the sentiment analysis of online monitoring new zealand phone number list specialist Clipit , which appears daily on the website of the Financieele Dagblad. Messages from more than 10,000 online sources, including the popular microblog Twitter, Hyves and Facebook are scanned every day and divided into 5 different categories: very positive, positive, neutral, negative and very negative. In this way, a public opinion is created for each fund. This is compared with the share price of the fund in question.