Oriented Trading Investor's Bond Rating

With the rapid development of China's credit bond market, many non-bank secured and unsecured bonds emerged, resulting in the diversification and complication of credit rating. This thesis first compares the theoretic applicability and empirical result of the popular quantitative rating methods,and introduce the probability of default measurement model used by commercial banks,finally the author choose the multiple Probit analysis method. In view of the public rating's lag defection and transactional investor's profiting mode, this thesis establishes the transactional investor-oriented rating model.This thesis references the altman's z-score model and rating agencies' rating variable to conclude the financial variables used by the model, and use some qualitative variables that are quant-measured to modify the quantitative rating. After fixing the explainary variables, this thesis adopts factor analysis and orthogonal transformation method to group the financial variables, which can reduce the number of explainary variables and eliminate the co-linearity among the explainary variables. Every single orthogonal factor represents certain ability of the issuer.Next, this thesis uses the public rating as explained variable and builds a bond rating model based on the sorted multiple Probit analysis, and modifies the quantitative rating qualitatively. The empirical test shows the internal rating approximate the public rating properly, but it deviates largely when rating the extreme-level bonds. Then this thesis uses yield spread and pricing model to validate the effectiveness of the internal raing. In view of the instability of the yield spread, this thesis investigate the relationship between yield spread and marco-economy to adjust the basic yield spread during different economic cycles.This thesis also tries to use market spread to deduce internal rating backward, which is proved invalid and financial variables can't affect the market spread. Although the quanlitative variables are significant, they can't explain the yield spread's change in time. This shows the china's bond market is ineffective, subsequent research should focus on locating the key factor to explain the market spread. For the further research, this thesis suggests to add more explainary variables, improve the econometric model, run multi-period robustness test and take the industry factor into consideration.