Modelling sovereign credit ratings: Neural networks versus ordered probit

Sovereign credit ratings are becoming increasingly important both within a financial regulatory context and as a necessary prerequisite for the development of emerging capital markets. Using a comprehensive dataset of rating agencies and countries over the period 1989-1999, this paper demonstrates that artificial neural networks (ANN) represent a superior technology for calibrating and predicting sovereign ratings relative to ordered probit modelling, which has been considered by the previous literature to be the most successful econometric approach. ANN have been applied to classification problems with great success over a wide range of applications where there is an absence of a precise theoretical model to underpin the relationships in the data. The results for sovereign credit ratings presented here corroborate other researchers' findings that ANN are highly effective classifiers.

[1]  John H. Aldrich,et al.  Linear probability, logit and probit models , 1984 .

[2]  Duarte Trigueiros,et al.  Neural networks and empirical research in accounting , 1996 .

[3]  Cheng Hsiao,et al.  Logit and Probit Models , 1992 .

[4]  Asad Zaman,et al.  Statistical Foundations for Econometric Techniques , 1996 .

[5]  A. Refenes.  Neural Networks in the Capital Markets , 1994 .

[6]  H. Daniels,et al.  Application of MLP Networks to Bond Rating and House Pricing , 1999, Neural Computing & Applications.

[7]  Ingoo Han,et al.  Ordinal Pairwise Partitioning (OPP) Approach to Neural Networks Training in Bond rating , 1997, Intell. Syst. Account. Finance Manag..

[8]  Tarun K. Sen,et al.  Predicting Bond Ratings Using Neural Networks: A Comparison with Logistic Regression , 1997, Intell. Syst. Account. Finance Manag..

[9]  Alvin J. Surkan,et al.  Neural networks for bond rating improved by multiple hidden layers , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[10]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[11]  Pietra Rivoli,et al.  Politics and Perceived Country Creditworthiness in International Banking , 1990 .

[12]  Melody Y. Kiang,et al.  Managerial Applications of Neural Networks: The Case of Bank Failure Predictions , 1992 .

[13]  Edward I. Altman,et al.  FINANCIAL RATIOS, DISCRIMINANT ANALYSIS AND THE PREDICTION OF CORPORATE BANKRUPTCY , 1968 .

[14]  Nelson C. Mark,et al.  The Relative Importance of Political and Economic Variables in Creditworthiness Ratings , 1998, SSRN Electronic Journal.

[15]  Ingoo Han,et al.  Case-based reasoning supported by genetic algorithms for corporate bond rating , 1999 .

[16]  Lourdes Treviño,et al.  Systematic Differences in the Determinants of Foreign Currency Sovereign Ratings by Rating Agency , 2000 .

[17]  John Moody,et al.  Architecture Selection Strategies for Neural Networks: Application to Corporate Bond Rating Predicti , 1995, NIPS 1995.

[18]  Ashok Vir Bhatia.  Sovereign Credit Ratings Methodology an Evaluation , 2002, SSRN Electronic Journal.

[19]  Lars Peter Hansen,et al.  Advances in Economics and Econometrics: Advances in Economics and Econometrics , 2003 .

[20]  V. Kvasnicka,et al.  Neural and Adaptive Systems: Fundamentals Through Simulations , 2001, IEEE Trans. Neural Networks.

[21]  David M. Kreps,et al.  Advances In Economics and Econometrics: Theory And Applications: Seventh World Congress , 1997 .

[22]  S. Dutta,et al.  Bond rating: a nonconservative application of neural networks , 1988, IEEE 1988 International Conference on Neural Networks.

[23]  Lourdes Treviño,et al.  Local vs Foreign Currency Ratings: What Determines Sovereign Transfer Risk? , 2001 .

[24]  S. Lee,et al.  Relative Importance of Political Instability and Economic Variables on Perceived Country Creditworthiness , 1993 .

[25]  Carmen M. Reinhart.  Default, Currency Crises and Sovereign Credit Ratings , 2002 .

[26]  Soumitra Dutta,et al.  Bond rating: A non-conservative application of neural networks , 1988 .

[27]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[28]  Richard Taffler,et al.  LDC Credit-risk Forecasting and Banker Judgement , 2001 .

[29]  Richard T. Redmond,et al.  Expert systems for bond rating: a comparative analysis of statistical, rule‐based and neural network systems , 1993 .

[30]  R. Taffler,et al.  The Assessment of Company Solvency and Performance using a Statistical Model , 1983 .

[31]  Selwyn Piramuthu,et al.  Financial credit-risk evaluation with neural and neurofuzzy systems , 1999, Eur. J. Oper. Res..

[32]  Jose C. Principe,et al.  Neural and adaptive systems : fundamentals through simulations , 2000 .

[33]  Kurt Hornik,et al.  Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks , 1990, Neural Networks.

[34]  J. Bessis,et al.  Risk Management in Banking , 1998 .

[35]  R. Kaplan,et al.  Statistical Models of Bond Ratings: A Methodological Inquiry , 1979 .

[36]  Lourdes Treviño,et al.  The Statistical Determinants of Local Currency Sovereign Ratings , 2000 .

[37]  Moshe Leshno,et al.  Neural network prediction analysis: The bankruptcy case , 1996, Neurocomputing.

[38]  Richard Cantor,et al.  Determinants and Impact of Sovereign Credit Ratings , 1996 .

[39]  Daniel Enache,et al.  Analyzing Credit Risk Data: A Comparison of Logistic Discrimination, Classification Tree Analysis, a , 1997 .

[40]  J. Cosset,et al.  The Determinants of Country Risk Ratings , 1991 .