Credit Risk Evaluation for Listed Company Based on Adaboost-LVQ

To protect the investors' interest and risk management of credit agency, conducting credit risk evaluation for listed company is very important. A new model based on Adaboost-LVQ neural network was proposed for credit risk evaluation. By defining credit risk as a listed company was special treated, the credit risk evaluation was changed into a pattern recognition problem and AdaBoost-LVQ neural network was introduced. 99 samples from the Shanghai and Shenzhen A-share stock market in 2004 to 2010 was used to test the classification effect of the new model. 70 in them were used as train set while the rest of them being used as test set. As a result, it was found that the new model had a good classification effect that 86.21% of test samples can be correctly classified.