基于BP神经网络的商业银行信用风险评估研究

发布时间：2019-01-03 11:36

【摘要】：随着经济全球化特别是金融全球化的快速发展,我国金融市场受到多方面因素的影响,不稳定性日趋明显,而我国目前正处于利率市场化的快速推进阶段,商业银行面临的机遇和挑战并存,特别是信用风险的挑战前所未有。当前商业银行的生存环境竞争日趋激烈,如何科学有效的管理信贷业务中存在的信用风险直接关系到商业银行的健康发展。处于改革转型和发展阶段的商业银行,原有的信用风险的管理体系已经难以适用,传统的分析方法无法满足其在新形势下快速发展的需求。基于此,本文立足于商业银行现阶段的基本特征,试图将神经网络研究方法运用到信贷业务中信用风险管理问题研究上,以期提供一种行之有效的风险评估技术。本文首先对现有的商业银行信用风险评估模型进行了梳理分析和论证,在界定商业银行信用风险内涵的基础上,较为深入的研究影响信用风险的各项因素,从而总结出现有信用风险管理体系的不足。进一步的,本文从五个层面上提取出17个指标,并将这一指标体系纳入到BP神经网络中,从而建立了一套完整的商业银行信用风险评估模型。最后,通过广泛的数据收集,应用MATLAB统计软件对模型的准确性进行研究和论证。最终结果表明,本文构建的风险评估模型具有较高的判别准确率,在实践中有助于商业银行对信贷业务的信用风险进行有效的评估,为信用风险管理提供了可靠的参考依据,具有一定的研究价值。
[Abstract]:With the rapid development of economic globalization, especially financial globalization, the financial market of our country is affected by many factors, and the instability is becoming more and more obvious. Commercial banks face both opportunities and challenges, especially the challenge of credit risk. At present, the living environment of commercial banks is becoming more and more competitive. How to manage credit risks scientifically and effectively is directly related to the healthy development of commercial banks. Commercial banks in the stage of reform, transformation and development, the original credit risk management system has been difficult to apply, the traditional analysis method can not meet its rapid development in the new situation. Based on this, based on the basic characteristics of commercial banks at the present stage, this paper attempts to apply the neural network research method to the study of credit risk management in credit operations, in order to provide an effective risk assessment technology. In this paper, the existing commercial bank credit risk assessment model is firstly analyzed and demonstrated. On the basis of defining the connotation of the commercial bank credit risk, the factors affecting the credit risk are deeply studied. In order to sum up the shortcomings of credit risk management system. Furthermore, this paper extracts 17 indexes from five levels, and brings this index system into BP neural network, thus establishing a complete credit risk assessment model of commercial banks. Finally, through extensive data collection, the accuracy of the model is studied and proved by using MATLAB statistical software. The final results show that the risk assessment model constructed in this paper has a high accuracy, which is helpful for commercial banks to evaluate the credit risk of credit business effectively, and provides a reliable reference for credit risk management. It has certain research value.
【学位授予单位】：内蒙古财经大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP183;F832.4

【参考文献】