Ordinal Pairwise Partitioning (OPP) Approach to Neural Networks Training in Bond rating

Abstract

Statistical classification methods such as multivariate discriminant analysis have been widely used in bond rating classification in spite of the limitations of the methodology. Recently, neural networks have emerged as new methods for business classification. This approach to neural networks training is to categorize a new instance as one of the predefined bond classes. Such a conventional approach has limitations in dealing with the ordinal nature of bond rating. In addition, most of the prior studies have used sample data which are evenly divided among the classes. However, the natural population in real application is usually unevenly divided among the classes. Under such circumstances, it is hard to achieve good predictive performance. As the number of classes to be recognized increases, the predictive performance decreases. In this article, to increase the predictive performance in real‐world bond rating, we propose the ordinal pairwise partitioning (OPP) approach to backpropagation neural networks training. The main idea of the OPP approach is to partition the data set in the ordinal and pairwise manner according to the output classes. Then, each backpropagation neural networks model is trained by using each partitioned data set and is separately used for classification. Experimental results show that the predictive performance of the proposed OPP approach can be significantly enhanced, when compared to the conventional neural networks modeling approach as well as multivariate discriminant analysis. The OPP approach has two computation methods, and we discuss under which circumstances one method performs better than the other. We also show the generalizability of the OPP approach. © 1997 by John Wiley & Sons, Ltd.

Ordinal Regression Methods: Survey and Experimental Study

“…The predictions are then combined by a union utility function. Finally, binary SVMs were also applied to ordinal regression [15], by making use of the ordinal pairwise partitioning approach [14]. This approach is composed of four different reformulations of the classical OneVsOne and OneVsAll paradigms.…”

“…Indeed, ordinal regression problems can be said to be between classification and regression, presenting some similarities and differences. Some of the fields where ordinal regression is found are medical research [5]- [11], age estimation [12], brain computer interface [13], credit rating [14]- [17], econometric modelling [18], face recognition [19]- [21], facial beauty assessment [22], image classification [23], wind speed prediction [24], social sciences [25], text classification [26], and more. All these works are examples of application of specifically designed ordinal regression models, where the ordering consideration improves their performance with respect to their nominal counterparts.…”

Determinants of Australian Listed Property Trust Bond Ratings

“…Several studies show that ANNs can be applied to bond rating: Dutta and Shekhar (1988); Surkan and Singleton (1990); Maher and Sen (1997); Kwon et al (1997) Maher and Sen (1997) compared the performance of neural networks with that of logistic regression. NN performed better than a traditional logistic regression model.…”

“…The best performance of the model was 70% (42 out of 60 samples). Kwon et al (1997) compared the predictive performance of ordinal pairwise partitioning approach to back propagation neural networks, conventional (CNN) modelling approach and MDA. They used 2365 Korean bond-rating data and demonstrated that NNs with OPP had the highest accuracy (71-73%), followed by CNN (66-67%) and MDA (58-61%).…”

“…Further, the reclassification of tranches into three classes could enhance model performance because mathematical and statistical approaches have general limits in dealing with the ordinal nature of bond rating. It is known that as the number of bond classifications increases, the predictive power could likely decrease (Kwon et al 1997). Table 3 provides summary statistics over time and by sector.…”