Credit rating analysis with support vector machines and neural networks: A market comparative study

Zan Huang, Hsinchun Chen, Chia Jung Hsu, Wun Hwa Chen, Soushan Wu

Research output: Contribution to journalArticle

546 Citations (Scopus)

Abstract

Corporate credit rating analysis has attracted lots of research interests in the literature. Recent studies have shown that Artificial Intelligence (AI) methods achieved better performance than traditional statistical methods. This article introduces a relatively new machine learning technique, support vector machines (SVM), to the problem in attempt to provide a model with better explanatory power. We used backpropagation neural network (BNN) as a benchmark and obtained prediction accuracy around 80% for both BNN and SVM methods for the United States and Taiwan markets. However, only slight improvement of SVM was observed. Another direction of the research is to improve the interpretability of the AI-based models. We applied recent research results in neural network model interpretation and obtained relative importance of the input financial variables from the neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and Taiwan markets.

Original languageEnglish (US)
Pages (from-to)543-558
Number of pages16
JournalDecision Support Systems
Volume37
Issue number4
DOIs
StatePublished - Sep 2004

Fingerprint

Support vector machines
Neural Networks (Computer)
Artificial Intelligence
Neural networks
Taiwan
Backpropagation
Research
Artificial intelligence
Benchmarking
Learning systems
Statistical methods
Support Vector Machine
Credit rating
Support vector machine
Comparative study
Credit
Neural Networks
Comparative Study
Rating
Back-propagation neural network

Keywords

  • Backpropagation neural networks
  • Bond rating prediction
  • Credit rating analysis
  • Cross-market analysis
  • Data mining
  • Input variable contribution analysis
  • Support vector machines

ASJC Scopus subject areas

  • Management Information Systems
  • Information Systems
  • Information Systems and Management

Cite this

Credit rating analysis with support vector machines and neural networks : A market comparative study. / Huang, Zan; Chen, Hsinchun; Hsu, Chia Jung; Chen, Wun Hwa; Wu, Soushan.

In: Decision Support Systems, Vol. 37, No. 4, 09.2004, p. 543-558.

Research output: Contribution to journalArticle

Huang, Zan ; Chen, Hsinchun ; Hsu, Chia Jung ; Chen, Wun Hwa ; Wu, Soushan. / Credit rating analysis with support vector machines and neural networks : A market comparative study. In: Decision Support Systems. 2004 ; Vol. 37, No. 4. pp. 543-558.
@article{2d85ec6b0fba4990845115ed9745e5a5,
title = "Credit rating analysis with support vector machines and neural networks: A market comparative study",
abstract = "Corporate credit rating analysis has attracted lots of research interests in the literature. Recent studies have shown that Artificial Intelligence (AI) methods achieved better performance than traditional statistical methods. This article introduces a relatively new machine learning technique, support vector machines (SVM), to the problem in attempt to provide a model with better explanatory power. We used backpropagation neural network (BNN) as a benchmark and obtained prediction accuracy around 80{\%} for both BNN and SVM methods for the United States and Taiwan markets. However, only slight improvement of SVM was observed. Another direction of the research is to improve the interpretability of the AI-based models. We applied recent research results in neural network model interpretation and obtained relative importance of the input financial variables from the neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and Taiwan markets.",
keywords = "Backpropagation neural networks, Bond rating prediction, Credit rating analysis, Cross-market analysis, Data mining, Input variable contribution analysis, Support vector machines",
author = "Zan Huang and Hsinchun Chen and Hsu, {Chia Jung} and Chen, {Wun Hwa} and Soushan Wu",
year = "2004",
month = "9",
doi = "10.1016/S0167-9236(03)00086-1",
language = "English (US)",
volume = "37",
pages = "543--558",
journal = "Decision Support Systems",
issn = "0167-9236",
publisher = "Elsevier",
number = "4",

}

TY - JOUR

T1 - Credit rating analysis with support vector machines and neural networks

T2 - A market comparative study

AU - Huang, Zan

AU - Chen, Hsinchun

AU - Hsu, Chia Jung

AU - Chen, Wun Hwa

AU - Wu, Soushan

PY - 2004/9

Y1 - 2004/9

N2 - Corporate credit rating analysis has attracted lots of research interests in the literature. Recent studies have shown that Artificial Intelligence (AI) methods achieved better performance than traditional statistical methods. This article introduces a relatively new machine learning technique, support vector machines (SVM), to the problem in attempt to provide a model with better explanatory power. We used backpropagation neural network (BNN) as a benchmark and obtained prediction accuracy around 80% for both BNN and SVM methods for the United States and Taiwan markets. However, only slight improvement of SVM was observed. Another direction of the research is to improve the interpretability of the AI-based models. We applied recent research results in neural network model interpretation and obtained relative importance of the input financial variables from the neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and Taiwan markets.

AB - Corporate credit rating analysis has attracted lots of research interests in the literature. Recent studies have shown that Artificial Intelligence (AI) methods achieved better performance than traditional statistical methods. This article introduces a relatively new machine learning technique, support vector machines (SVM), to the problem in attempt to provide a model with better explanatory power. We used backpropagation neural network (BNN) as a benchmark and obtained prediction accuracy around 80% for both BNN and SVM methods for the United States and Taiwan markets. However, only slight improvement of SVM was observed. Another direction of the research is to improve the interpretability of the AI-based models. We applied recent research results in neural network model interpretation and obtained relative importance of the input financial variables from the neural network models. Based on these results, we conducted a market comparative analysis on the differences of determining factors in the United States and Taiwan markets.

KW - Backpropagation neural networks

KW - Bond rating prediction

KW - Credit rating analysis

KW - Cross-market analysis

KW - Data mining

KW - Input variable contribution analysis

KW - Support vector machines

UR - http://www.scopus.com/inward/record.url?scp=2442665617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2442665617&partnerID=8YFLogxK

U2 - 10.1016/S0167-9236(03)00086-1

DO - 10.1016/S0167-9236(03)00086-1

M3 - Article

AN - SCOPUS:2442665617

VL - 37

SP - 543

EP - 558

JO - Decision Support Systems

JF - Decision Support Systems

SN - 0167-9236

IS - 4

ER -