Forecasting credit rating by Using Data Mining Techniques in Taiwan Banking Industry

• Main Authors: Yen-ming Huang, 黃彥銘
• Other Authors: Shian-chang Huang
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/05246079425535928948

碩士 === 國立彰化師範大學 === 企業管理學系 === 100 === During the financial tsunami of 2008, many derivatives or financial institution have problems, credit rating grade was lowered, fairness of credit rating company, has been questioned. In recent years the rapid development of world economy, banks play the role of the funding agency who become increasingly important in the economy , With the evolution of the financial innovation of new financial instruments continued to introduce new.Banking range caused by the bank increased business risk, and increase the difficulty and complexity of the credit rating and financial supervision; for financial institutions to establish a set of objective credit rating model indeed is necessary. In this study, as the Chinese credit rating company, the promulgation of the Bank class is divided into four groups, the variable reference CAMELS evaluation indicators to take into account the financial variables in addition to join the Corporate Governance and Macroeconomic Variables, research methods J48 with AdaBoostM1, and MultiBoost algorithms compare models the correct rate, to identify the most accurately forecast the credit rating model. The empirical results found that the J48 binding AdaBoostM1 established credit rating model classification accuracy rate of 94%.