Research On Credit Risk Identification Model Of Small And Medium - Sized Enterprises In Commercial Banks

This thesis mainly focuses on the research of commercial bank credit risk identification problem for SMEs. The goal of this thesis is to study the establishment of model system which can effectively identify the potential credit risk of SMEs.This paper through the study of literature has established the feasibility and the overall structure. Then it defined SMEs and elaborated the standard of definition. It analysed the connotation, category, status quo and causes of SMEs credit risk. This thesis analysed commercial bank credit risk evaluation system of traditional and modern commercial bank credit risk evaluation system. Put forward the significance of establishing the evaluation system of SMEs credit risk to commercial banks. Then this paper analyzed the relevant evaluation index by qualitative analysis method and initially identified the 14 indexes including financial and non-financial indicators. And this paper, through the method of quantitative analysis have graded on non-financial evaluation index. This paper tested the significant difference of the preliminary establishing evaluation index by the independent sample T test method and completed the further screening.finally it selected 8 significant evaluation indexes. Finally, based on the BP neural network, this paper constructed the commercial bank credit risk identification model using the neural network toolbox of MATLAB. And through the introduction of training samples of the above model.this paper found the law of identifing the enterprise credit risk.Finally the accuracy of testing model in the recognition of enterprise credit risk can reach 75%. Model constructed in this paper, compared with other similar studies, in the aspect selection of credit risk evaluation index, have certain advantages.This paper had a full consideration on the effects of non financial indicators to the small and medium-size enterprise credit risk evaluation.And through the significant difference test method,this paper had a screening and optimization of the all index.However, this paper have some limitations,in the aspect of the evaluation results, which need to be improved.I hope this paper can make some contribution to the future research on commercial bank credit risk for SMEs in terms of recognition.