Two Essays on Information Asymmetry

• Other Authors: Lahtinen, Kyre (authoraut)
• Format: Others
• Language: English; English
• Published: Florida State University
• Subjects: Finance; Management
• Online Access: http://purl.flvc.org/fsu/fd/FSU_migr_etd-8829

I examine the effect of information in two distinct settings. First, using a unique sample of Twitter posts, also called tweets, I examine the impact of information on social media on the return, volume, and volatility of the stock market using word list and algorithmic content analysis. I show market returns may be predicted using confidence and sentiment levels. Volume is best predicted by confidence. Volatility is most related to sentiment. I examine one dimension of Twitter user characteristics, namely gender. My results show that men are more confident and less optimistic than women when they communicate about stocks. I find differences in the ability of communications by men and women to predict market returns, volume, and volatility. Second, I assess the validity of the dividends signaling hypothesis when accounting for asymmetric information. I find that firms that have higher levels of asymmetric information, both in absolute levels and in relative levels to the firm, are less likely to pay or increase dividends. Dividend changes are not strong predictors of future earnings performance when firms are sorted by asymmetric information. Positive market reactions to dividend increases or initiations are linked to other forces. I do not find support for the dividend signaling hypothesis. === A Dissertation submitted to the Department of Finance in partial fulfillment of the requirements for the degree of Doctor of Philosophy. === Spring Semester, 2014. === April 9, 2014. === Dividends, Sentiment, Stock Market, Twitter === Includes bibliographical references. === Bong Soo Lee, Professor Directing Thesis; Thomas Zuehlke, University Representative; Don Autore, Committee Member; Yi Zhou, Committee Member.