An empirical analysis of structuring commercial mortgage-backed securities credit ratings : Australian evidence

The ultimate goal of structuring Commercial Mortgage-Backed Securities (CMBS) transactions is to obtain a high credit rating as this has an impact on the yield obtainable and the success of the issue. Though bond rating agencies claim that their ratings reflect each agency’s opinion about an issue’s potential default risk and rely heavily on a committee’s analysis of the issuer’s ability and willingness to repay its debt and therefore researchers would not be able to replicate their ratings quantitatively (Kim 2005), we follow previous researchers who gone ahead and replicated bond ratings on the premise that the financial variables extracted from public financial statements, such as financial ratios, contain a large amount of information about a company’s credit risk (Huang et al. 2004). We use artificial neural networks (ANN) and ordinal regression (OR) as alternative methods to predict CMBS ratings. OR results show that rating agencies use only a subset of variables they describe or indicate as important to CMBS rating as some of the variables they use were statistically insignificant. Overall, ANN show superior results to OR in predicting CMBS ratings.