An Analysis of the Consumer Credit Risk in Loans- The Applicaton of Neural Network

• Main Authors: Jung-Lung Hung, 洪榮隆
• Other Authors: Li-Hua Lai
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/43283075292397468227

碩士 === 國立高雄第一科技大學 === 風險管理與保險所 === 91 === An Analysis of the Consumer Credit Risk in Loans The Application of Neural Network Student：Jung-Lung Hung Advisor：Dr. Li-Hua Lai Department of Risk Management and Insurance National Kaohsiung First University of Science and Technology Abstract Credit risk measurement is the techniques that help organization decide whether or not grant credit to consumer who apply to them. This article surveys a sample of 240 consumer loans which had been classified as either good or bad accounts which were measured via 14 predictor variables, and using back-propagation network (BPN) of artificial neural networks to analyze database, then it outperforms 90% predictive correct rate, and the research finding are: 1.The several risk factors (such as nrphones, residence, nrquest, credit history, time with present employer, occupation, income, gender ) have significant impact on claim and it have become establishes as major tools in forecasting financial risk in consumer lending. 2.A correct risk prediction model which built by artificial neural networks has also been developed. The model may assist organizations’ administrators using their applicant’s demographics for proving risk management.