This research established the combined model of credit risk and credit Rating that could be used to predict the risk of fiduciary loan, base on the rules established by the Basel Committee on Banking Supervision. Stepwise Logistic Regression was used to predict the probability of customer default successfully. The credit score was predicted, as well as the credit rating was divided to 10 groups by cluster analysis method. In addition, this research also explicitly defines the interval value of each credit rating level. Finally we give each credit rating level a practical connotation of risk.