Abstract
In this paper we investigate the ability of a number of different ordered probit models to predict ratings based on firm-specific data on business and financial risks. We investigate models which are based on momentum, drift and ageing, and compare them with alternatives which take the initial rating of the firm and its previous actual rating into account. Using data on US bond issuing firms, as rated by Fitch, over the years 2000 to 2007, we compare the performances of these models for predicting the ratings both in-sample and out-of-sample using root mean squared errors, Diebold-Mariano tests of forecast performance and contingency tables. We conclude that both initial and previous states have a substantial influence on rating prediction.
Highlights
It is well known that ratings agencies provide an independent assessment of the risk of a counterparty using information on the balance sheet, the profit and loss account and private information on the management of the entity, summarized using a rating scale running from the highest rating AAA to the lowest CCC
When we examine the predictive ability in- and out-of-sample using measures of the root mean squared error, the Diebold & Mariano (1995) prediction test and evaluate the proportion of correct predictions using Merton’s correct prediction statistic (Merton (1981)) we find that the state dependence model is better on this measure than alternatives
The resulting model shows that non-linearities and state dependence terms improve the fit of a model seeking to determine a firm’s credit rating
Summary
It is well known that ratings agencies provide an independent assessment of the risk of a counterparty using information on the balance sheet, the profit and loss account and private information on the management of the entity, summarized using a rating scale running from the highest rating AAA to the lowest CCC. Examination of ratings behavior over time by Blume et al (1998) documented that credit ratings, on average, became worse as increased volatility in corporate creditworthiness during the mid-1980s and early 1990s was accompanied by downward momentum in credit ratings. Ratings can deteriorate because firms have lower credit quality, if they becoming more leveraged for example, and later studies by Amato & Furfine (2004) identified no secular change in rating standards in data from 1984-2001. Rather, their results implied that ratings changes were driven by changes to business and financial risks, not by cycle-related changes to rating standards. The large number of rating downgrades during the US corporate credit meltdown in 2001–2002 and 2007-9 casts some doubt on the extent to which ratings see through the cycle
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
