基于模糊积分支持向量机集成的商业银行信用风险评估模型研究

The Model of Credit Risk Assessment in Commercial Banks on Fuzzy Integral Support Vector Machines Ensemble

信用风险管理一直是银行和其他金融机构最关心的问题之一。随着我国经济体制的改革深入、市场机制的建立与完善以及资本市场、银行业的迅速发展,现行的信用评估体制与方法赶不上经济改革发展的需要。本文建立了基于模糊积分的支持向量机集成方法,该方法综合考虑了子支持向量机的输出重要性。用此方法对商业银行的信用风险进行评估,并与单个支持向量机和最多投票原则的支持向量机集成进行比较,实证结果表明,本文提出的方法具有更高的分类精度,证实了该方法的可行性和有效性,为银行建立一套更加可靠的评估系统提供了理论依据。

Assessment of credit risk is very important to commercial bank and other financial organization. With the development of consumer credit in domestic commercial banks, the existing risk assessment methods can＇t satisfy the need. A SVMs ensemble method based on fuzzy integral is presented to develop the assessment system in this paper. This method aggregates the outputs of separate component SVMs with the importance of each component SVM. The empirical results demonstrate that the proposed method outperforms a single SVM and traditional SVMs aggregation technique via majority voting in terms of classification accuracy. It provides the theory for developing the system for risk assessment of commercial bank.