Textual analysis, information diffusion, and asset returns

Text sentiment;Textual analysis;Earnings announcements;Soft information

Chebonenko, Tatiana

关键词: Text sentiment;    Textual analysis;    Earnings announcements;    Soft information;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/46826/Tatiana_Chebonenko.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

The past decade has seen the rapid development of different techniques to retrieve additional information from financial news. Various software and methods of news analytics are used to quantify textual information. Text sentiment extracts text’s attitude by counting negative words and has proved extremely useful in a variety of contexts. The literature interprets it in three ways: quantitative information, soft news, and psychological sentiment. In the first chapter, we use a quasi-natural experiment to show that text sentiment reflects primarily omitted quantitative information and does not capture soft news or sentiment. We first extract text sentiment from earnings call transcripts with dictionary and supervised-learning methods, and then compare how it predicts returns during overnight and intraday calls; specifically, whether text sentiment explains a larger portion of stock returns for overnight calls. The overnight and intraday cases differ only in the timing of a quarterly report. Overnight calls are dominated by quantitative news from a quarterly report, while the intraday cases contain mostly soft news and sentiment. Text sentiment explains overnight returns well but fails to predict returns or volatility during intraday calls. Thus, text sentiment reflects news only during periods dominated by quantitative information.Using textual analysis, the second chapter provides evidence that earnings announcements contain mostly company-specific news and very little industry news. The main problem is how to estimate the causal effect of an earnings announcement on firm and industry stock returns given that many other factors could affect both returns. I address the identification problem by directly measuring earnings news with text sentiment extracted from earnings call transcripts. I find that text sentiment explains a large portion of announcer stock returns but very little of industry returns during earnings announcements. The result is robust to alternative measures of text sentiment and industry classification.The third chapter finds that an earnings announcement emits information to option traders. I use textual analysis to separate the earnings news flow to equity and option markets. Text sentiment predicts stock price crash risk and volatility spike option estimates which illustrates separate earnings information flow to the option traders. I find that firm size and analyst coverage explain most of the predictability for volatility spikes. At the same time, text sentiment proves to be useful in prediction of sudden drops in stock price.