商业银行信用风险混合判别模型及实证分析[J].docVIP

商业银行信用风险混合判别模型 及实证分析 焦继文1，2，王福重2，郭春媛1 （1. 山东大学管理学院，山东济南，250100； 2.北京航空航天大学管理学院，北京，100083） 摘要：随着世界金融一体化趋势和金融市场波动性的加剧，商业银行信用风险管理已成为国内外政府、金融机构乃至企业关注的焦点问题。因此，对商业银行信用风险进行有效监管就显得非常迫切。本文以上海证券交易所上市的山东板块的24家公司作为样本，通过选取相对合理的指标体系，使用主成分分析对指标数据信息进行有效的压缩，进而构建了主成分分析与神经网络技术相结合的混合评估判别模型。实证结果表明，该种混合模型的识别误差很小且效率较高，为商业银行有效地评估和识别信用风险提供了一种可行的方法。 关键词：商业银行；信用风险；主成分分析；人工神经网络；混合评估判别模型 中图分类号：F830.33 文献标识码：A Mixed Assessing and Forecasting Model of the Commercial Bank Credit Risks and Empirical Analysis JIAO Ji-wen1，2WANG Fu-zhong3，GUO Chun-yuan2 （1. School of Management, China University of Mining and Technology, Xuzhou 221008, P.R. China; 2. School of Management, Shandong University, Jinan 250100, P.R. China; 3. School of Economics and Management, Beihang University, Beijing 10083, P.R. China） Abstract: Along with the world financial integration tendency and financial market fluctuation intensifying, the commercial bank credit risk management has become the focal point which the domestic and foreign governments, the financial institutions and even the enterprises pay attention to. Therefore, it appears extremely urgent to carries on effective supervising and managing to the commercial bank credit risks. This article takes the 24 companies in Shandong province as the sample from the Shanghai Stock Exchange, through selecting relatively reasonable index system, uses principal components analysis compressing the index messages effectively, and then has made up the mixed assessing and recognizing model in which the principal component analysis and the neural network technique are integrated. Empirical results show that this kind of mixed model’s recognition errors are very small and the efficiency is very high. It has provided a feasible method for the commercial bank credit risk assessing and recognizing. Key words commercial bank; credit risk; principal components analysis; Artificial neural network; mixed assessing and recognizing model 1. 引言 商业银行信用风险的管理一直是国内外金融界关注的焦点问题。《新巴塞尔协议》的实施无