Updating a credit-scoring model based on new attributes without realization of actual data

Yong Han Ju, So Young Sohn

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

Funding small and medium-sized enterprises (SMEs) to support technological innovation is critical for national competitiveness. Technology credit scoring models are required for the selection of appropriate funding beneficiaries. Typically, a technology credit-scoring model consists of several attributes and new models must be derived every time these attributes are updated. However, it is not feasible to develop new models until sufficient historical evaluation data based on these new attributes will have accumulated. In order to resolve this limitation, we suggest the framework to update the technology credit scoring model. This framework consists of ways to construct new technology credit-scoring model by comparing alternative scenarios for various relationships between existing and new attributes based on explanatory factor analysis, analysis of variance, and logistic regression. Our approach can contribute to find the optimal scenario for updating a scoring model.

Original languageEnglish
Pages (from-to)119-126
Number of pages8
JournalEuropean Journal of Operational Research
Volume234
Issue number1
DOIs
Publication statusPublished - 2014 Apr 1

Fingerprint

Credit Scoring
Updating
Attribute
Model-based
Model
Small and Medium-sized Enterprises
Scenarios
Competitiveness
Analysis of variance
Factor analysis
Factor Analysis
Logistic Regression
Analysis of variance (ANOVA)
Credit scoring
Scoring
Logistics
Resolve
Innovation
Update
Sufficient

All Science Journal Classification (ASJC) codes

  • Management Science and Operations Research
  • Modelling and Simulation
  • Information Systems and Management

Cite this

@article{048e60dbfe0449eaa8bb81d5d4df1497,
title = "Updating a credit-scoring model based on new attributes without realization of actual data",
abstract = "Funding small and medium-sized enterprises (SMEs) to support technological innovation is critical for national competitiveness. Technology credit scoring models are required for the selection of appropriate funding beneficiaries. Typically, a technology credit-scoring model consists of several attributes and new models must be derived every time these attributes are updated. However, it is not feasible to develop new models until sufficient historical evaluation data based on these new attributes will have accumulated. In order to resolve this limitation, we suggest the framework to update the technology credit scoring model. This framework consists of ways to construct new technology credit-scoring model by comparing alternative scenarios for various relationships between existing and new attributes based on explanatory factor analysis, analysis of variance, and logistic regression. Our approach can contribute to find the optimal scenario for updating a scoring model.",
author = "Ju, {Yong Han} and Sohn, {So Young}",
year = "2014",
month = "4",
day = "1",
doi = "10.1016/j.ejor.2013.02.030",
language = "English",
volume = "234",
pages = "119--126",
journal = "European Journal of Operational Research",
issn = "0377-2217",
publisher = "Elsevier",
number = "1",

}

Updating a credit-scoring model based on new attributes without realization of actual data. / Ju, Yong Han; Sohn, So Young.

In: European Journal of Operational Research, Vol. 234, No. 1, 01.04.2014, p. 119-126.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Updating a credit-scoring model based on new attributes without realization of actual data

AU - Ju, Yong Han

AU - Sohn, So Young

PY - 2014/4/1

Y1 - 2014/4/1

N2 - Funding small and medium-sized enterprises (SMEs) to support technological innovation is critical for national competitiveness. Technology credit scoring models are required for the selection of appropriate funding beneficiaries. Typically, a technology credit-scoring model consists of several attributes and new models must be derived every time these attributes are updated. However, it is not feasible to develop new models until sufficient historical evaluation data based on these new attributes will have accumulated. In order to resolve this limitation, we suggest the framework to update the technology credit scoring model. This framework consists of ways to construct new technology credit-scoring model by comparing alternative scenarios for various relationships between existing and new attributes based on explanatory factor analysis, analysis of variance, and logistic regression. Our approach can contribute to find the optimal scenario for updating a scoring model.

AB - Funding small and medium-sized enterprises (SMEs) to support technological innovation is critical for national competitiveness. Technology credit scoring models are required for the selection of appropriate funding beneficiaries. Typically, a technology credit-scoring model consists of several attributes and new models must be derived every time these attributes are updated. However, it is not feasible to develop new models until sufficient historical evaluation data based on these new attributes will have accumulated. In order to resolve this limitation, we suggest the framework to update the technology credit scoring model. This framework consists of ways to construct new technology credit-scoring model by comparing alternative scenarios for various relationships between existing and new attributes based on explanatory factor analysis, analysis of variance, and logistic regression. Our approach can contribute to find the optimal scenario for updating a scoring model.

UR - http://www.scopus.com/inward/record.url?scp=84890425656&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890425656&partnerID=8YFLogxK

U2 - 10.1016/j.ejor.2013.02.030

DO - 10.1016/j.ejor.2013.02.030

M3 - Article

VL - 234

SP - 119

EP - 126

JO - European Journal of Operational Research

JF - European Journal of Operational Research

SN - 0377-2217

IS - 1

ER -