Bio-Inspired Credit Risk Analysis

by Yu, Lean; Wang, Shouyang; Lai, Kin Keung; Zhou, Ligang

• ISBN13: 9783540778028
• ISBN10: 3540778020
• Format: Hardcover
• Copyright: 2008-06-05
• Publisher: Springer Verlag

Summary

Credit risk analysis is one of the most important topics in the field of financial risk management. Due to recent financial crises and regulatory concern of Basel II, credit risk analysis has been the major focus of financial and banking industry. Especially for some credit-granting institutions such as commercial banks and credit companies, the ability to discriminate good customers from bad ones is crucial. The need for reliable quantitative models that predict defaults accurately is imperative so that the interested parties can take either preventive or corrective action. Hence credit risk analysis becomes very important for sustainability and profit of enterprises. In such backgrounds, this book tries to integrate recent emerging support vector machines and other computational intelligence techniques that replicate the principles of bio-inspired information processing to create some innovative methodologies for credit risk analysis and to provide decision support information for interested parties.

Table of Contents

Credit Risk Analysis with Computational Intelligence: An Analytical Survey	p. 1
Credit Risk Analysis with Computational Intelligence: A Review	p. 3
Introduction	p. 3
Literature Collection	p. 5
Literature Investigation and Analysis	p. 7
What is Credit Risk Evaluation Problem?	p. 8
Typical Techniques for Credit Risk Analysis	p. 8
Comparisons of Models	p. 17
Implications on Valuable Research Topics	p. 23
Conclusions	p. 24
Unitary SVM Models with Optimal Parameter Selection for Credit Risk Evaluation	p. 25
Credit Risk Assessment Using a Nearest-Point-Algorithm-basedSVM with Design of Experiment for Parameter Selection	p. 27
Introduction	p. 27
SVM with Nearest Point Algorithm	p. 29
DOE-based Parameter Selection for SVM with NPA	p. 33
Experimental Analysis	p. 35
Conclusions	p. 38
Credit Risk Evaluation Using SVM with Direct Search for Parameter Selection	p. 41
Introduction	p. 41
Methodology Description	p. 43
Brief Review of LSSVM	p. 43
Direct Search for Parameter Selection	p. 45
Experimental Study	p. 47
Research Data	p. 47
Parameter Selection with Genetic Algorithm	p. 48
Parameters Selection with Grid Search	p. 49
Experimental Results	p. 50
Conclusions	p. 54
Hybridizing SVM and Other Computational Intelligent Techniquesfor Credit Risk Analysis	p. 57
Hybridizing Rough Sets and SVM for Credit Risk Evaluation	p. 59
Introduction	p. 59
Preliminaries of Rough Sets and SVM	p. 61
Basic Concepts of Rough Sets	p. 61
Basic Ideas of Support Vector Machines	p. 62
Proposed Hybrid Intelligent Mining System	p. 63
General Framework of Hybrid Intelligent Mining System	p. 63
2D-Reductions by Rough Sets	p. 64
Feature Selection by SVM	p. 65
Rule Generation by Rough Sets	p. 66
General Procedure of the Hybrid Intelligent Mining System	p. 67
Experiment Study	p. 68
Corporation Credit Dataset	p. 69
Consumer Credit Dataset	p. 70
Concluding Remarks	p. 72
A Least Squares Fuzzy SVM Approach to Credit Risk Assessment	p. 73
Introduction	p. 73
Least Squares Fuzzy SVM	p. 74
SVM	p. 74
FSVM	p. 77
Least Squares FSVM	p. 79
Experiment Analysis	p. 81
Conclusions	p. 84
Evaluating Credit Risk with a Bilateral-Weighted Fuzzy SVMModel	p. 85
Introduction	p. 85
Formulation of the Bilateral-Weighted Fuzzy SVM Model	p. 89
Bilateral-Weighting Errors	p. 89
Formulation Process of the Bilateral-weighted fuzzy SVM	p. 91
Generating Membership	p. 93
Empirical Analysis	p. 95
Dataset 1: UK Case	p. 96
Dataset 2: Japanese Case	p. 98
Dataset 3: England Case	p. 100
Conclusions	p. 102
Evolving Least Squares SVM for Credit Risk Analysis	p. 105
Introduction	p. 105
SVM and LSSVM	p. 108
Evolving LSSVM Learning Paradigm	p. 111
General Framework of Evolving LSSVM Learning Method	p. 111
GA-based Input Features Evolution	p. 113
GA-based Parameters Evolution	p. 117
Research Data and Comparable Models	p. 119
Research Data	p. 119
Overview of Other Comparable Classification Models	p. 121
Experimental Results	p. 123
Empirical Analysis of GA-based Input Features Evolution	p. 123
Empirical Analysis of GA-based Parameters Optimization	p. 126
Comparisons with Other Classification Models	p. 129
Conclusions	p. 131
SVM Ensemble Learning for Credit Risk Analysis	p. 133
Credit Risk Evaluation Using a Multistage SVM Ensemble Learning Approach	p. 135
Introduction	p. 135
Previous Studies	p. 138
Formulation of SVM Ensemble Learning Paradigm	p. 140
Partitioning Original Data Set	p. 140
Creating Diverse Neural Network Classifiers	p. 142
SVM Learning and Confidence Value Generation	p. 143
Selecting Appropriate Ensemble Members	p. 144
Reliability Value Transformation	p. 146
Integrating Multiple Classifiers into an Ensemble Output	p. 146
Empirical Analysis	p. 148
Consumer Credit Risk Assessment	p. 149
Corporation Credit Risk Assessment	p. 151
Conclusions	p. 154
Credit Risk Analysis with a SVM-based Metamodeling Ensemble Approach	p. 157
Introduction	p. 157
SVM-based Metamodeling Process	p. 160
A Generic Metalearning Process	p. 160
An Extended Metalearning Process	p. 163
SVM-based Metamodeling Process	p. 165
Experimental Analyses	p. 173
Research Data and Experiment Design	p. 173
Experimental Results	p. 174
Conclusions	p. 177
An Evolutionary-Programming-Based Knowledge Ensemble Modelfor Business Credit Risk Analysis	p. 179
Introduction	p. 179
EP-Based Knowledge Ensemble Methodology	p. 181
Brief Introduction of Individual Data Mining Models	p. 182
Knowledge Ensemble based on Individual Mining Results	p. 185
Research Data and Experiment Design	p. 188
Experiment Results	p. 189
Results of Individual Models	p. 189
Identification Performance of the Knowledge Ensemble	p. 191
Identification Performance Comparisons	p. 193
Conclusions	p. 195
An Intelligent-Agent-Based Multicriteria Fuzzy Group DecisionMaking Model for Credit Risk Analysis	p. 197
Introduction	p. 197
Methodology Formulation	p. 201
Experimental Study	p. 206
An Illustrative Numerical Example	p. 206
Empirical Comparisons with Different Credit Datasets	p. 208
Conclusions and Future Directions	p. 221
References	p. 223
Subject Index	p. 239
Biographies of Four Authors of the Book	p. 243
Table of Contents provided by Publisher. All Rights Reserved.