摘要
随着互联网金融的快速发展,供应链金融、P2P、众筹、电子商务众多创新融资模式的相继出现,大大缓解了我国中小企业普遍融资难的问题,但随之而来另一突出问题是企业信用违约事件屡屡发生。由于我国中小企业大部分为非上市公司,信息不对称,造成信用风险监管的问题日益突出。Moody’s KMV公司开发了期权定价的PFM模型,对解决非上市公司信用风险监管难题提供了有效的途径。但PFM模型对非线性样本数据适用准确性差,估计结果不理想。文章采用数据挖掘中的支持向量机(SVM)回归分析方法,利用其适用小样本、高维性和非线性数据分析的特点,对PFM模型在我国非上市公司风险度量进行了实证研究。结果表明此方法的运用,使商业银行可以准确地对非上市公司信用风险进行度量,进而优化选择信贷决策,同时对PFM模型在我国信用风险度量方法的研究方面提供了一定的理论参考依据。
With the rapid development of Internet finance, supply chain finance, P2 P, crowdfunding and electronic commercehave emerged, greatly relieving the universal financial difficulty of small and medium-sized enterprises in our country.However, the credit default of enterprises frequently occurs. Since most small and medium-sized enterprises in our country are non-listed and the information is asymmetric, the supervision of credit risk becomes more and more serious. The option pricing PFM model developed by Moody＇s KMV provides an effective way to solve the credit risk of non-listed companies. Based on this model, the paper applies Support Vector Machine（SVM）regression analysis to measure the credit risk of non-listed companies in China and conducts empirical research. The results show that,with this method, commercial banks can accurately measure the non-listed companies＇ credit risk and optimize credit decision, thus providing theoretical reference for the PFM model in related research.
作者
刘艳春
崔永生
LIU Yanchun CUI Yongsheng(College of Business Administration, Liaoning University, Shenyang 110036, China)
出处
《辽宁大学学报（哲学社会科学版）》
2016年第6期88-97,共10页
Journal of Liaoning University(Philosophy and Social Sciences Edition)
基金
辽宁经济社会发展立项课题"大数据时代中小企业融资创新模式研究"(2016lslktziglx-11)
