A Study on the Corporate Governance and Financial Variables Using Bayesian Network

The purpose of this study is to analyze characteristics between financial variables of a company that have influence of corporate governance structure. This study attempts to overcome an uniformed linear assumption about independent and dependent variables that are used in the previous studies. In other words, it focuses on finding causality between components of financial variables of a firm that have influence on corporate governance. Transparency of corporate governance can be improved by providing accounting information users or supervisory institutions with a variety of information on how a financial variable has influence on corporate governance through direct or indirect casual relationship with other financial variables. For the purpose, this study proposes Baysian Network and presents Makov Blanket drawn from the Baysian Network. According to the research on 2,232 firm-years whose corporate governance was evaluated and disclosed by Corporate Governance Service from 2012 to 2015, it is empirically demonstrated that the methodology used in this study present statistically significant results for corporate governance and financial variables.