An artificial intelligence approach to predicting bond ratings
Author: Dutta, Soumitra ; Shekar, ShashiINSEAD Area: Technology and Operations Management In: Expert systems in economics, banking and management - Pau, Louis-François - 1989 - Book Language: EnglishDescription: p. 59-68.Type of document: INSEAD ChapterNote: Please ask us for this itemAbstract: The default risk of most bonds are rated by various independent organizations, eg Moody's and SandP. These ratings are often used by investors to define allowable bond purchases and to measure the risk characteristics of investments in bonds. It is not known what model, if any, these rating agencies use for rating the various bond issues. Past researchers have used multi-variate regression models to relate the corporate bond ratings with the financial ratios of the corporation, but have had limited successes. In this paper, results are presented from an initial attempt at predicting bond ratings by using the neural network model from the domain of artificial intelligence and comparing the results with those obtained by using regression models.Item type | Current location | Collection | Call number | Status | Date due | Barcode | Item holds |
---|---|---|---|---|---|---|---|
![]() |
Digital Library | Available | BC000476 |
Ask Qualtrics
The default risk of most bonds are rated by various independent organizations, eg Moody's and SandP. These ratings are often used by investors to define allowable bond purchases and to measure the risk characteristics of investments in bonds. It is not known what model, if any, these rating agencies use for rating the various bond issues. Past researchers have used multi-variate regression models to relate the corporate bond ratings with the financial ratios of the corporation, but have had limited successes. In this paper, results are presented from an initial attempt at predicting bond ratings by using the neural network model from the domain of artificial intelligence and comparing the results with those obtained by using regression models.
Digitized
There are no comments for this item.