• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>The 19th International Conference on Information Quality,Big Data: management amp; Data quality >Modeling Corporate Credit rating under Basel II: Data Mining Application
【24h】

Modeling Corporate Credit rating under Basel II: Data Mining Application

机译：在巴塞尔协议II：数据挖掘应用程序中建模企业信用评级

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

The main purpose of this study was using data mining techniques to build corporate credit rating modelsrnin accordance with the Basel II and guidelines of FSC. Data was observed from 2003 to 2005 onrncorporate financial statements combined with variables of credit record from bank A database.rnTechnologies such as Microsoft SQL Server 2012 Analysis Services have made credit rating a new arearnwhere bank can gain a competitive advantage in their corporate loaning business. Particularly throughrndata mining extraction of hidden predictive information from large databases for the bank to predictrncredit default probability from the developed model. Neural networks, decision trees and logisticrnregression techniques were adapted to establish predictive model. To compare the proportion of samplingrnerror, results found that the best model was training by the ratio of 1: 2. The performance of logisticrnregression models has better predictive power compared with neural networks and decision treernapproaches. Financial information variables in the model had more predictive power to explainrncorporate default. This model remains certain stability and predictive validity when applied in differentrntime period.
机译：这项研究的主要目的是根据巴塞尔协议II和FSC指南，使用数据挖掘技术来建立公司信用评级模型。我们观察到了2003年至2005年的企业财务报表以及银行A数据库中信用记录变量的数据。诸如Microsoft SQL Server 2012 Analysis Services之类的技术已使信用评级成为一种新的领域，银行可以在其企业贷款业务中获得竞争优势。特别是通过数据挖掘，从大型数据库中提取隐藏的预测信息，以供银行从已开发的模型中预测信用违约概率。采用神经网络，决策树和逻辑回归技术建立预测模型。为了比较抽样误差的比例，结果发现最好的模型是按1：2的比率进行训练。与神经网络和决策树方法相比，逻辑回归模型的性能具有更好的预测能力。该模型中的财务信息变量具有更多的预测力来解释企业违约。当在不同时间段应用时，该模型保持一定的稳定性和预测有效性。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号