The classification approach, previously considered in credit card scor-7 ing, is extended to multi class classification in application to credit rating of bonds. 8 The classification problem is formulated as minimization of a penalty constructed 9 with quadratic separating functions. The optimization is reduced to a linear pro-10 gramming problem for finding optimal coefficients of the separating functions. Var-11 ious model constraints are considered to adjust model flexibility and to avoid data 12 overfitting. The classification procedure includes two phases. In phase one, the clas-13 sification rules are developed based on "in-sample" dataset. In phase two, the classi-14 fication rules are validated with "out-of-sample" dataset. The considered methodol-15 ogy has several advantages including simplicity in implementation and classification 16 robustness. The algorithm can be applied to small and large datasets. Although the 17 approach was validated with a finance application, it is quite general and can be 18 applied in other engineering areas. 19
CITATION STYLE
Bugera, V., Uryasev, S., & Zrazhevsky, G. (2008). Classification Using Optimization: Application to Credit Ratings of Bonds. In Computational Methods in Financial Engineering (pp. 211–237). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-540-77958-2_11
Mendeley helps you to discover research relevant for your work.