Credit Assessment Model of Non-linear Combining Forecast for Individual Housing Loan Based on GP

摘要

Aiming at the low predictive accuracies of single statistical models, this paper presents a combining forecast model for credit assessment in individual housing loan. Based on the two single statistical models of linear regression and logistic regression, this paper constructed a non-linear combining forecast model by using genetic programming (GP) to search a non-linear function and used the model for individual housing loan in commercial banks. The application results indicate that the non-linear combining forecast model based on GP increases the predictive accuracy effectively. Compared with the two single statistical models, the predictive accuracy is increased by 3.40% and 2.83% and the type II error rate is decreased by 10.48% and 8.46% respectively, which is more significant for commercial banks to keep away from credit risks in individual housing loan.