Faculty of Economics and Business Administration Publications Database

Analyzing the Relationship between Differentiated Online Sentiment and Company-Specific Stock Prices

Akolk, Fabian
Beck, Roman
Link External Source: Online Version
Year: 2015

Practitioners and researchers alike increasingly use social media messages as an additional source of information when dealing with stocks. Based on emotion theory and an established sentiment lexicon, we develop and apply an open source dictionary for the analysis of seven different emotions in 5.5 million Twitter messages on 33 S&P 100 companies. We find varying explanatory power of different emotions (esp. happiness and depression) for company-specific stock price movements over a period of three months.