News-Induced Dynamic Networks for Market Signaling: Understanding the Impact of News on Firm Equity Value

• 成果类型: Article
• 署名作者: Chen, Kun; Li, Xin; Luo, Peng; Zhao, J. Leon
• 署名单位: Southern University of Science & Technology; City University of Hong Kong; Sichuan University; The Chinese University of Hong Kong, Shenzhen
• 刊物名称: INFORMATION SYSTEMS RESEARCH
• ISSN/ISSBN: 1047-7047
• DOI: 10.1287/isre.2020.0969
• 发表日期: 2021
• 页码: 356-377
• 关键词: BAD-NEWS; information-content; investor attention; Social media; STOCK; Sentiment; strategy; performance; returns; predictability

摘要

Firm relations, which inform the competitive environments of firms, are critical to firm operations and are often factored into investor decisions. Previous studies and practices have considered relatively stable and long-term business relations as an indicator of firm market value. Public news often reports on business relations, especially dynamic and short-term opportunities and challenges involving different partners. Learning about firm relations from news is commonly done by human investors but has not been studied systematically in previous research, leading to a research opportunity in market signaling via dynamic firm relations. In this study, we propose a new approach to market signaling by leveraging text-mining methods to extract cobenefit/counter-benefit networks based on firms' mutual or conflicting interests in business events. We find that the resulting networks in the long term are partially aligned with firm cooperation and competition relationships, confirming their semantic implications for investor perception and attention. Our empirical study further shows that dynamic networks formed in a short time period (measured by firm centrality in networks) have significant impacts on firm equity value, after controlling for market activities and other information from news, such as volume, sentiment, and comentions. We show that dynamic networks can provide additional value in predicting firm equity value over stable networks. Moreover, the negative effects of counter-benefit networks emerge rapidly and persist longer than the positive effects of cobenefit networks. This study provides new insights into investor perception of news and suggests new research directions for financial text mining. Our research findings on market signaling via news-induced networks also have an impact on financial practices, such as market analysis and automatic trading.