• Journal of the Korean Data and Information Science Society
• /
• 제24권1호
• /
• pp.73-83
• /
• 2013

신용평가방법에서 등급의 계량화 중 신용등급 변화 검정방법은 등급별로 추정된 예측부도율과 실제부도율과의 동질성을 검정하는 방법으로 한 시점에 대한 이항검정과 카이제곱검정 등이 있고, 여러시점의 정확성을 검증하는 방법으로 정규성검정, 확장된 신호등검정 등이 있다. 본 연구에서는 현실적인 상황을 고려하여 이런 검정방법들이 상관관계가 존재하는 경우에 등급별 동질성 검정방법을 소개하고 이 방법들을 신용평가 이외에 다양한 분야의 자료에 활용할 수 있음을 알아본다.

• 경영정보학연구
• /
• 제1권1호
• /
• pp.51-61
• /
• 1999

• 한국컴퓨터정보학회논문지
• /
• 제29권5호
• /
• pp.11-20
• /
• 2024

최근 IT 기술의 발전과 더불어 금융 시장에서의 불확실성이 증대되는 상황에서 기업 신용등급 평가의 중요성을 인식하고, 이를 개선하기 위한 새로운 접근 방식으로 딥러닝 모델인 TabNet을 제안한다. 이에 본 연구에서는 TabNet을 활용하여 기업 신용등급을 예측하고, 이의 예측 성능을 기존 머신러닝 방법론과 상세하게 비교한다. 한국의 주요 증권시장에 상장된 기업들의 재무 데이터를 기반으로 TabNet 알고리즘을 적용하여 신용등급 예측 모델을 구축하고, 다양한 머신러닝 모델과의 성능을 비교 분석하였다. 실험 결과, TabNet 모델은 Precision 0.884, F1이 0.895로 기존의 머신러닝 모델들보다 우수한 성능을 보였으며, 고위험 기업을 저위험 기업으로 잘못 분류하는 경우가 다른 머신러닝 모델보다 적어 TabNet의 우수성을 확인하였다. 이는 TabNet이 기업 신용등급 예측에 있어 효과적인 도구로 활용될 수 있으며, 금융기관의 신용 위험 관리 및 의사 결정 과정을 지원할 수 있을 것으로 기대한다.

• 한국콘텐츠학회논문지
• /
• 제16권3호
• /
• pp.81-90
• /
• 2016

본 연구는 당기 주가수익률과 차기 신용등급 및 신용등급 변화와의 관련성을 검증하는 것을 목적으로 한다. 신용등급평가사들은 개별 기업의 채무불이행위험(default risk)을 측정하여 최종 신용등급을 결정하는데 기업의 높은 주가수익률은 낮은 위험(default risk)으로 인지될 가능성이 있다. 반면 시장참여자들은 효율적으로 높은 수익을 달성하기 위하여 규모가 크고 안정적인 기업보다 고수익을 달성할 수 있는 신용위험(risk)이 높은 기업들의 주식을 선호할 가능성 역시 배제할 수 없다. 이는 실증적으로 해결되어야 할 문제이며 현재까지 이러한 관련성을 고찰한 연구는 부재하다. 본 연구는 2002년부터 2013년까지 회사채를 발행한 유가증권 상장기업을 대상으로 당기 주가수익률과 차기 신용등급 및 신용등급의 관련성을 검증하였고, 그 결과를 요약하면 다음과 같다. 먼저 당기 주가수익률은 차기 신용등급과 유의한 음(-)의 관련성이 있는 것으로 나타났다. 이는 신용평가사들이 주가수익률을 채무불이행 위험의 대리변수로 고려하지 않음을 예측케 하는 결과이고, 오히려 투자자들은 신용등급이 낮은 기업의 주식을 선호한다고 해석할 수 있다. 본 연구는 직관과는 달리 주가수익률과 신용등급의 음(-)의 관련성을 찾은 최초의 연구로써 신용평가사 및 시장참여자들에게 의미 있는 통찰력을 제공할 것으로 기대한다.

• Journal of Information Processing Systems
• /
• 제11권2호
• /
• pp.173-183
• /
• 2015

At present, it is simple to the electronic commerce credit scoring model, as a brush credit phenomenon in E-commerce has emerged. This phenomenon affects the judgment of consumers and hinders the rapid development of E-commerce. In this paper, that E-commerce credit evaluation model that uses a Gaussian density function is put forward by density test and the analysis for the anomalies of E-commerce credit rating, it can be fond out the abnormal point in credit scoring, these points were calculated by nonlinear credit scoring algorithm, thus it can effectively improve the current E-commerce credit score, and enhance the accuracy of E-commerce credit score.

• 한국정보시스템학회지:정보시스템연구
• /
• 제30권4호
• /
• pp.203-226
• /
• 2021

Purpose The purpose of this study is to enhance the accuracy score of personal credit scoring using the convolutional neural networks and secure the transparency of the deep learning model using eXplainalbe Artifical Inteligence(XAI) technique. Design/methodology/approach This study built a classification model by using the convolutional neural networks(CNN) and applied a methodology that is transformation of numerical data to imaged data to apply CNN on personal credit data. Then layer-wise relevance propagation(LRP) was applied to model we constructed to find what variables are more influenced to the output value. Findings According to the empirical analysis result, this study confirmed that accuracy score by model using CNN is highest among other models using logistic regression, neural networks, and support vector machines. In addition, With the LRP that is one of the technique of XAI, variables that have a great influence on calculating the output value for each observation could be found.

• 국제학술발표논문집
• /
• The 6th International Conference on Construction Engineering and Project Management
• /
• pp.122-126
• /
• 2015

Credit enhancement is absolutely essential for financing Independent Power Producer (IPP) projects in Asia particularly for countries whose sovereign credit rating is on non-investment grade and foreign investment is difficult to achieve. Due to nexus of agreements among varies parties in IPP project, it is hard to clearly visualize the roles of these agreements. Examples are: What credit enhancement factors are most influential to minimize the associated risks of IPP projects? Why are they powerful? What are their roles? Who are less powerful and what are the obstacles that causes them less powerful? A research is conducted to identify the credit enhancement factors for IPP projects in Asia. IPP professionals validated 27 out of 28 identified credit enhancement factors, and five factor groupings were made through factor analysis. Afterwards, network theory is applied to find the unanswered questions, which by graphical and mathematical representations show that the host government's credit enhancement, MDBs, ECAs and other parties' credit enhancement are prominent and of great importance to handle the associated risks of IPP projects in Asia

• 수산경영론집
• /
• 제47권2호
• /
• pp.43-59
• /
• 2016

This study investigated the difference of the effects of public loan programs in fishery industry on management performance from a balanced score card (BSC) perspective depending on the type of loan, scale of fund, period of support and business category, using the financial data of fisheries firms having the balance of loan at the end of 2014. The key factors influencing credit rating change were also analyzed after public loan support. From a integrative perspective, results show that the firms supported by working fund have higher management performance than the firms supported by facility fund. The firms received large scale fund showed higher management performance than the firms received small scale fund. While management performance was decreasing or slowing down over time after financial support, management performance of the firms supported by facility fund improved over time. From a non-financial perspective, the firms received facility fund invested more in education and growing perspective than the firms received working fund. As the size of fund increased, the investment in education, growing, internal process and customer increased. Personnel expenses and employee benefits for education and growing has increased over time. However, the firms with facility fund restricted the expenses of education, personnel expenses and employee benefits as time goes by. Because the effects of public loan on credit rating of fisheries corporations have no statistical significance, it has become known that the financial support of public loan program has no influence on the change of credit rating of fisheries corporations. This study attempted performance analysis from a BSC perspective which combine factors of non-financial perspective with factors of financial perspective. Findings from this study suggest the direction of microscopic performance analysis of public loan in fishery industry.

• The Journal of Asian Finance, Economics and Business
• /
• 제3권3호
• /
• pp.5-15
• /
• 2016

This study examines whether Korean rating agencies such as Korea Investors Service (KIS), National Information & Credit Evaluation (NICE), and Korea Ratings Corporation (KR), incorporate corporate governance into their corporate bond ratings in Korea. We find that the Korean rating agencies assign higher ratings to the bonds issued by Chaebol (Korean business group) affiliated firms. Our results also indicate that those rating agencies give higher ratings to the bonds with greater foreign investor share ownership. Moreover, if the rating agencies value corporate governance, higher rated firms should issue bonds at lower yield to maturity. We discover that Chaebol affiliation is counted favorably by the rating agencies. We find that investors are willing to pay lower risk premium for bonds with higher institutional ownership, but higher risk premium to bonds with greater equity ownership in the form of depository receipts. Therefore, even if the rating agencies and investors in Korea consider corporate governance (Chaebol affiliation and ownership structure) an important determinant in bond ratings and the yields to maturity, they have opposite views on institutional ownership and share ownership in the form of depository receipts.

• 가정과삶의질연구
• /
• 제27권6호
• /
• pp.95-109
• /
• 2009

This study focused on the development of a scale to measure the ability of consumers to use credit cards. The purposes of this study were to develop a tool which would be able to measure consumer knowledge, consumer skills and consumer attitudes. Data were collected from 313 credit card using consumers and were analyzed by employing a goodness of fit test, principal component analysis & confirmatory factor analysis(Amos 5.0), multiple regression. The results from this study were as follows: 1) Six factors of consumer knowledge(16-items) were identified: damage salvation; credit delinquency; personal credit information; credit provision period; credit & credit card issuance; credit delinquent striking out a record & credit rating. The total variance was 55.86%. 2) Three factors of consumer skills(17-items) were identified: credit delinquency & over-consumption; credit card management; and loss & damage salvation. The total variance was 62.90%. 3) Three factors of consumer attitudes(16-items) were identified: credit delinquency & credit; credit card issuance & use; and credit card management. The total variance was 58.75%.