基于支持向量机方法的中小企业信用评级问题研究

发布时间：2018-09-07 12:58

【摘要】：中小企业在我国国民经济和社会发展中处于重要的战略地位,但中小企业长期以来一直面临着融资困境,制约了其健康发展。形成融资难的原因是多方面的,如果能建立一套适用中小企业特点的信用评级方法,将在很大程度上解决银企之间的信息不对称问题,从而缓解其融资困境。但是,我国目前中小企业信用评级基本沿用大型企业的评级方法,致使其评级结果并不能准确反映真实的信用水平,难以真实反映中小企业信用风险。近年来,中小企业信用评级研究方兴未艾,取得了一系列的研究成果,也积累了很多成功经验,但仍然缺乏一套能充分反映中小企业自身特点的专用信用评级系统,因此需要对已有的中小企业信用评价指标体系和技术路线予以优化,为中小企业健康发展提供金融支持服务。本研究致力于中小企业信用评级指标体系的选取和技术路线的优化。文章首先对信用评级的相关理论和问题进行了详细的分析和研究,然后通过对目前商业银行指标体系缺陷性的研究情况下,结合中小企业自身特点以及信用状况水平进行系统分析后,建立了一套针对中小企业自身的信用评级指标体系。为了克服现有基于传统统计模型评级方法的局限性,本研究力图将信用评级转换为模式识别和聚类分析,通过选用一种小样本学习理论支持向量机(SVM)方法对中小企业信用状况进行评估,形成较为先进的中小企业信用评级方法。文章对该方法进行了详细的介绍,最后通过实证分析并与BP神经网络进行比较最终证明了该方法的有效性,并对未来的深入研究进行了展望。
[Abstract]:Small and medium-sized enterprises (SMEs) are in an important strategic position in the national economy and social development of our country, but SMEs have been facing financing difficulties for a long time, which has restricted their healthy development. There are many reasons for the difficulty in financing. If we can establish a set of credit rating methods suitable for the characteristics of small and medium-sized enterprises, we will solve the problem of information asymmetry between banks and enterprises to a great extent, and then alleviate their financing difficulties. However, at present, the credit rating of small and medium-sized enterprises in our country basically follows the method of large enterprises, which makes the result of credit rating can not accurately reflect the true credit level, and it is difficult to truly reflect the credit risk of small and medium-sized enterprises. In recent years, the research on the credit rating of small and medium-sized enterprises is in the ascendant, and has made a series of research results and accumulated a lot of successful experiences. However, there is still a lack of a special credit rating system that can fully reflect the characteristics of small and medium-sized enterprises. Therefore, it is necessary to optimize the existing credit evaluation index system and technical route of SMEs to provide financial support services for the healthy development of SMEs. This study is devoted to the selection of credit rating index system and the optimization of technical route. Firstly, this paper makes a detailed analysis and research on the relevant theories and problems of credit rating, and then through the current research on the defect of the index system of commercial banks, After analyzing the characteristics and credit status of SMEs, a set of credit rating index system is established. In order to overcome the limitations of traditional statistical model based rating methods, this study attempts to transform credit rating into pattern recognition and clustering analysis. By selecting a small sample learning theory support vector machine (SVM) method to evaluate the credit status of small and medium-sized enterprises, a more advanced credit rating method for small and medium-sized enterprises is formed. This paper introduces the method in detail, and finally proves the effectiveness of this method by empirical analysis and comparison with BP neural network, and looks forward to further research in the future.
【学位授予单位】：安徽财经大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：F275;F832.4;TP181

【参考文献】