论文部分内容阅读
本文利用国内某商业银行的中小企业贷款违约数据,采用Logistic模型和随机效应Logistic模型对非上市中小企业贷款违约风险的影响因素进行实证分析。结果表明:非上市中小企业的硬信息指标特别是营运资金比率、负债比率对贷款违约有较大影响;软信息指标中的企业特征特别是企业在人行的信用记录、经营稳定性、水电税费变化趋势和银行信用对违约风险有显著影响;非上市中小企业企业主的个人特征,例如受教育程度、家庭年收入、持股比例和资产抵押率等对违约风险也具有较大影响。预测对比研究发现,软信息指标对贷款违约的预测能力高于硬信息指标,而随机效应Logistic模型的表现要强于Logistic模型;综合使用软信息和硬信息指标建立随机效应Logistic模型具有最佳的预测效果。研究结论对商业银行开展中小企业贷款业务和信贷风险评估具有参考价值。
In this paper, we use the data of SME loan defaults in a commercial bank in China and use the Logistic model and the random-effects Logistic model to analyze the influencing factors of the default risk of non-listed SMEs loans. The results show that the hard information indicators of non-listed SMEs, especially the working capital ratio and the debt ratio, have a great impact on the default of loans. The characteristics of soft information indicators, especially the credit records of enterprises, the stability of operation, The trend of change and bank credit have a significant impact on the default risk. The personal characteristics of non-listed SME business owners, such as education level, annual household income, shareholding ratio and asset mortgage rate, also have a significant impact on the default risk. Compared with the Logistic model, the performance of the random information logistic model is better than that of the hard information indicator. The combination of soft information and hard information indicators to establish the random effects Logistic model has the best prediction effect. The conclusion of the study is of reference value for commercial banks to carry out SME loan business and credit risk assessment.