A note comparing support vector machines and ordered choice models’ predictions of international banks’ ratings

Bellotti, Tony; Matoušek, Roman; Stewart, Chris

doi:10.1016/j.dss.2011.03.008

Cited by 47 publications

(19 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…external rating process such as linear regression (Horrigan 1996;West 1970), linear discriminant analysis (Pinch and Mingo 1973, 1 975), logit and probit (Altman and Katz 1976;Jackson and Boyd 1988) ordered logit and ordered probit (Kamstra et al 2001;Altman and Rijken 2004;Amato and Furfine 2004;Alejandro and Analía 2008;Bellotti et al 2011b), artifical intelligence techniques (Dutta and Shekhar 1988;Surkan and Singleton 1990;Kim et al 1993;Kwon et al 1997). Kim (2005), Huang et al (2004 and Lee (2007) show that artificial intelligence techniques (particularly neural networks and support vector machines) do not provide superior predictions of bond ratings compared with standard ordered-choice methods.…”

Section: Many Methodologies Have Been Developed In Recent Years Whichmentioning

confidence: 99%

Asymmetric Benchmarking in Bank Credit Rating

2012

View full text Add to dashboard Cite

This study proposes an information asymmetry hypothesis to examine why bank credit ratings vary among countries even when bank financial ratios remain constant. Countries are divided among those with low and high information asymmetry. The former include high-income countries, those in North America and West Europe regions, and those with strong institutional environment quality, whereas the latter group possess the opposite characteristics. This study hypothesizes that the influences of financial ratios on ratings are enhanced in low information asymmetry countries but reduced in countries with high information asymmetry. The sample includes the long-term credit ratings issued by Standard and Poor's from 86 countries during 2002-2008. The estimated results show that the effects of financial ratios on ratings are significantly affected by information asymmetries. Countries wishing to improve the credit ratings of their banks thus should reduce information asymmetry.

show abstract

Section: Many Methodologies Have Been Developed In Recent Years Whichmentioning

confidence: 99%

Asymmetric Benchmarking in Bank Credit Rating

2012

View full text Add to dashboard Cite

show abstract

“…Many methodologies have been developed in recent years which analyze the external rating process such as linear regression (Horrigan, 1966;West, 1970), linear discriminant analysis (Pinches and Mingo, 1973, 1975), logit and probit (Altman and Katz, 1976Jackson and Boyd, 1988) ordered logit and ordered probit (Kamstra et al, 2001;Altman and Rijken, 2004;Amato and Furfine, 2004;Alejandro and Analía, 2008;Bellotti et al, 2011b), artificial intelligence techniques (Dutta and Shekhar, 1988;Surkan and Singleton, 1990;Kim et al, 1993;Kwon et al, 1997). Kim (2005), Huang et al (2004) and Lee (2007) show that artificial intelligence techniques (particularly neural networks and support vector machines) do not provide superior predictions of bond ratings compared with standard ordered-choice methods.…”

Section: Credit Rating Determinantsmentioning

confidence: 99%

Asymmetric benchmarking in bank credit rating

Shen

Huang

Hasan

2012

Journal of International Financial Markets, Institutions and Mo

View full text Add to dashboard Cite

a b s t r a c tThis study proposes an information asymmetry hypothesis to examine why bank credit ratings vary among countries even when bank financial ratios remain constant. Countries are divided among those with low and high information asymmetry. The former include high-income countries, those in North America and West Europe regions, and those with strong institutional environment quality, whereas the latter group possess the opposite characteristics. This study hypothesizes that the influences of financial ratios on ratings are enhanced in low information asymmetry countries but reduced in countries with high information asymmetry. The sample includes the long-term credit ratings issued by Standard and Poor's from 86 countries during 2002-2008. The estimated results show that the effects of financial ratios on ratings are significantly affected by information asymmetries. Countries wishing to improve the credit ratings of their banks thus should reduce information asymmetry.

show abstract

“…Bellotti and Crook (2009) compared LR and SVM, but without showing the feature selection method for the LR. Bellotti, Matousek, and Stewarti (2011) compared LR with SVM, but for regression purposes, not for classification. They found that the SVM model outperforms LR.…”

Section: Introductionmentioning

confidence: 99%

An innovative feature selection method for support vector machines and its test on the estimation of the credit risk of default

Sariev

Germano

2018

Rev Financ Econ

View full text Add to dashboard Cite

Support vector machines (SVM) have been extensively used for classification problems in many areas such as gene, text and image recognition. However, SVM have been rarely used to estimate the probability of default (PD) in credit risk. In this paper, we advocate the application of SVM, rather than the popular logistic regression (LR) method, for the estimation of both corporate and retail PD. Our results indicate that most of the time SVM outperforms LR in terms of classification accuracy for the corporate and retail segments. We propose a new wrapper feature selection based on maximizing the distance of the support vectors from the separating hyperplane and apply it to identify the main PD drivers. We used three datasets to test the PD estimation, containing (1) retail obligors from Germany, (2) corporate obligors from Eastern Europe, and (3) corporate obligors from Poland. Total assets, total liabilities, and sales are identified as frequent default drivers for the corporate datasets, whereas current account status and duration of the current account are frequent default drivers for the retail dataset.

show abstract

A note comparing support vector machines and ordered choice models’ predictions of international banks’ ratings

Abstract: We find that Support Vector Machines virtually always predict international bank ratings better than ordered choice models.

Cited by 47 publications

References 33 publications

Asymmetric Benchmarking in Bank Credit Rating

Asymmetric Benchmarking in Bank Credit Rating

Asymmetric benchmarking in bank credit rating

An innovative feature selection method for support vector machines and its test on the estimation of the credit risk of default

Contact Info

Product

Resources

About