| Credit scoring plays a critical role in many areas such as business,finance,engineering and health.How to establish a simple and effective credit scoring method is an important topic in the field of credit scoring.There are often some redundant variables in the credit scoring model,so it is necessary to apply variable selection to give better results,which leads to another important topic,that is,model selection method based on credit scoring method,which aims to give the criteria of variable selection for a specific credit scoring method.This paper aims to develop a model selection method of a new credit scoring method,and give the theoretical properties.The KolmogorovSmirnov statistic is a commonly used criterion for evaluating credit scoring method’s performance.Directly Maximizes the Kolmogorov-Smirnov(DMKS)is a credit scoring method that optimizes KS statistic as an objective function for the first time.Like other credit scoring methods,DMKS method also has the problem of variable redundancy.To overcome this issue,we develop a new model selection method for choosing a suitable credit scoring model.We further prove the asymptotic optimality of our proposed method,that is,the KS statistics corresponding to the parameters obtained by the model selection method are asymptotically optimal compared to the KS statistics corresponding to all parameter estimates in the candidate models.In addition,a model selection method based on cross validation often has a large amount of calculation and runs slowly.Iterative Marginal Optimization(IMO)algorithm is used to accelerate the calculation of the proposed model selection criterion,so that our proposal can be applied when the sample size is large.In addition,the calculation of our proposed model selection method is further reduced by using the idea of the forward variable selection method,and thus the speed of selecting a suitable credit scoring model is accelerated.Simulation studies and a real data analysis show the effectiveness of the proposed model selection method. |
