Personal Commercial Micro-loan Defaults Prediction for Credit Risk Using:__An Auto-associative Neural Network-based Methodology

Traditionally,credit scoring aimed at distinguishing good payers from bad payers at the time of the application.Empirical micro loan default prediction models have been proposed and widely used in the last decades or so.A major problem is the imbalance of data,i.e.much more solvent data than default data.We propose auto-associative neural networks (AANN) that learns the identity mapping of input.By training the network,we built a Personal commercial micro-loan defaults predictor.More importantly,the methodology used to construct and validate models does not require advanced statistical analysis.
micro-loan credit risk auto-associative neural network
Yingli Huang Zhaowen Qiu
School of Economics and management,Northeast Forestry University,P.R.China,150040 School of Business,University of Surrey,Surrey,GU2 7XH,United Kingdom P.R.China,150040
国际会议
长沙
英文
823-828
2008-10-28(万方平台首次上网日期,不代表论文的发表时间)