Differential Emotions and the Stock Market - The Case of Company-specific Trading

摘要: Practitioners and researchers alike increasingly use social media messages as an additional source of information to analyse stock price movements. In this regard, previous preliminary findings demonstrate the incremental value considering multi-dimensional structure human emotions in sentiment analysis instead predominant assessment binary positive-negative valence emotions. Therefore, based on emotion theory established lexicon, we develop apply open dictionary for seven different (affection, happiness, satisfaction, fear, anger, depression, contempt).To investigate connection between differential movements approximately 5.5 million Twitter 33 S&P 100 companies their respective NYSE prices from Yahoo!Finance over a period three months. Subsequently, conduct lagged fixed-effects panel regression daily closing differences. The results generally support assumption necessity more differentiated sentiment. Moreover, comparing positive negative valence, find that only average emotionality strength has significant with company-specific specific reveals increase depression happiness isassociated decrease prices.