In view of such characteristics of Chinese corporate bonds as variance in industries’terms’types, unexpected fluctuation in rate of return and trading volume with its discontinuity and uncertainty, etc., an inquiry has been made into the method for calculating the downside risk of corporate bonds, followed by an analysis of the relationship between the risk and the returns. Based on the research and analysis of the characteristics of Chinese corporate bonds, the principles of bond screening and data processing can be determined, with the conditional VaR and ES models to be established by using the historical simulation method, the effectiveness of the models to be tested and compared by using the failure frequency method, and the relationship between risk and return to be analyzed according to the downside risk of the segmented portfolio investment. The Research results show that the VaR and ES models meet the requirements for an effective measurement of the downside risk under different confidence levels, with the ES model failure ratio being relatively low. Meanwhile, the results of the Chinese corporate bond segmentation portfolio show that the average return of each group is positively correlated with ES.