An Empirical Study on Predicting Illegal Disclosure of Listed Companies in China: from the Perspectives of Finance, Market and Governance
- 管理学院－已发表论文 
作为公开原则的核心体现,信息披露无疑是规范上市公司行为、保护中小投资者利益和提高证券市场配置效率的最有效手段。然而,我国一些上市公司无视法纪,信息披露违规屡禁不止,严重地违反了诚信准则。本文基于财务、市场和治理视角,运用条件Logistic回归模型对公司信息披露违规进行预警研究,实证结果表明:公司治理信息有助于提高预警模型的判别成功率,揭示上市公司的信息披露违规风险。而且,在违规前一年,基于财务指标、市场指标和治理指标的预警模型可以有效地提前甄别信息披露违规的上市公司。Information disclosure, as the core of the principle of openness,is the most effective means to regulate listed companies, protect minority shareholders and promote market efficiency to allocate resources. Unfortunately, some listed companies in China always violate the securities laws and regulations via illegal disclosure,and they even refuse to mend their ways despite of repeated admonition. This paper tries to develop the prediction model by applying Conditional Logistic regression from the perspectives of finance, market and governance.We find that corporate governance indicators play a key role in improving the accuracy of prediction model.Moreover, the Conditional Logistic prediction model, based on the financial indicators, market indicators and governance indicators, is able to effectively identify the companies likely to disclose information illegally one year before they do so.