If you have several other independent variables, you can try to use multiple dummy variables. For instance, one variable for AAA, one for AA, and one for A. The omitted variable is for these below A. However, adding too many dummy variables is not desirable when the number of independent variables is small.
h*k
12 楼
她不是说CMU嘛.
【在 r****x 的大作中提到】 : which uni?
k*h
13 楼
I guess he is talking about credit rating as an indepedent variable, instead of dependent variable, as he referred he is tring to examine the relationship between bond price and credit rating. Iidoteque is right about the ordinal nature of credit rating. But how to define credit rating depends on the nature of your research question. Many publications did code it as a cardinal variable. You can also probably treat them as a set of dummy variables, although you lose some information by treating