• 한국산업정보학회논문지
• /
• 제29권2호
• /
• pp.119-131
• /
• 2024

2019년 금융위원회의 온라인 대출 중개 서비스 도입 허용에 따라 핀다와 토스같은 핀테크 대출중개 플랫폼들이 가파르게 성장하였다. 하지만 대출중개 플랫폼으로 대출을 받은 고객들은 기존의 모집법인을 통해 대출을 받은 고객들보다 채무불이행 위험도가 더 높은 것으로 보고되었다. 본 연구는 생존분석 기법을 통해 신용대출을 받기 위해 대출중개 플랫폼을 통해 대출을 받은 고객들과 모집법인을 통해 대출을 받은 고객들의 채무불이행 위험도를 비교하고 각 대출중개 플랫폼별로 고객들의 채무불이행에 영향을 주는 특성들을 비교하였다. 분석을 위해 국내 캐피털사의 고객 데이터를 활용하였으며 카플란-마이어 분석 및 AFT 모형을 활용하였다. 모집법인을 통해 대출을 받은 고객들에 비해 대출중개 플랫폼을 통해 대출을 받은 고객들은 개인사업자보다 일반 대출자의 비중이 높고 부동산을 소유하고 있을 확률이 높았으며 대출진행기간이 더 짧았다. 또한 대출진행기간 중앙값(385일)을 기준으로 채무불이행 발생 비율이 더 높았다. AFT 모형을 통해 채무불이행 발생 시기를 분석한 결과 대출중개 플랫폼은 모두 모집법인에 비해 채무불이행 발생 확률이 높은 것으로 나타났다. 또한 대출중개 플랫폼을 통해 대출을 받은 고객의 특성들이 채무불이행 발생 시기에 주는 영향은 플랫폼별로 상이하게 나타났다. 이 결과는 대출중개 플랫폼별로 고객의 리스크 차이를 고려한 맞춤형 전략이 필요함을 보여준다.

• 한국전자거래학회지
• /
• 제23권3호
• /
• pp.207-224
• /
• 2018

온라인 P2P 대출(Online Peer-to-Peer Lending)이란 대출자(차입자)들이 인터넷 및 모바일 P2P 플랫폼을 통해 대출을 신청하면 P2P 플랫폼 기업이 이를 심사하고, 공개하여 불특정 다수가 자금을 빌려주고 이자를 받는 대출중개 서비스를 말한다. 국내외적으로 P2P 대출시장의 성장과 수익률에 대한 관심이 커진 상황에서 현재는 P2P 대출에 대한 안정성 측면에서 문제가 제기되고 있다. P2P 대출시장은 높은 수익률을 제공하지만 P2P 업체의 연체율과 부실률(채무불이행률)도 함께 높아지고 있는 실정이다. P2P 금융시장의 신뢰도를 높이기 위해서는 P2P 대출의 연체율과 채무불이행률을 줄이는 것이 무엇보다 중요하다. 본 연구는 세계적인 P2P 기업인 렌딩클럽(Lending Club)의 P2P 대출거래데이터베이스를 이용하여 인공지능기반의 P2P 채무불이행 예측모형을 구축하고자 한다. 구체적으로 벤치마크(benchmark) 모형으로 통계기법인 판별분석과 로지스틱 회귀분석을 이용하고, 인공지능기법으로는 신경망, CART, 그리고 C5.0을 이용하여 P2P 대출거래의 채무불이행 예측모형을 구축하고자 한다. 연구결과, P2P 대출거래의 채무불이행 예측을 위해 우선 고려해야 할 변수는 대출이자율이며, 중요도 3순위에 가장 많이 언급된 대출금액과 총부채상환비율도 고려해야 할 요인으로 추출되었다. 전통적인 통계기법보다는 인공지능기법의 예측성과가 더 좋은 것으로 나타났으며, 신경망의 경우 모든 데이터 셋에서 오분류율이 가장 낮은 예측모형으로 나타났다.

• Journal of the Korean Data and Information Science Society
• /
• 제26권6호
• /
• pp.1417-1426
• /
• 2015

본 연구는 한국장학재단의 2012-2014년간 일반 학자금 대출 자료를 활용하여 부실채권 보유 및 신용유의자로 분류될 수 있는 위험요인들을 파악하고, 부실 고위험군 예측모형을 개발했다. 예측모형 개발은 데이터마이닝 방법 중 의사결정나무 분석을 적용하였으며, 분석 패키지는 SAS Enterprise Miner 13.2를 활용했다. 개발된 모형은 25가지의 그룹으로 세분화 했으며, 부실 위험군에 영향을 미치는 주요 요인은 소득분위, 국가장학금 수혜유무, 나이, 연체계좌 보유 이력, 대학구분 (학부/대학원), 전공 계열, 월평균 상환액이 주요 요인으로 나타났다. 본 연구에서 개발된 부실 고위험군 예측모형은 장기연체로 인한 부실채권 발생 및 신용유의자 발생 예방을 위한 세분화된 관리서비스 제공을 위한 기초자료가 될 수 있을 것이다.

• KDI Journal of Economic Policy
• /
• 제44권3호
• /
• pp.1-26
• /
• 2022

This paper analyzes the effects of the cut in the legal maximum interest rate (from 27.4% to 24%) that occurred in February of 2018 on loan interest rates, the default rates, and the loan approval rate of borrowers in the non-banking sector. We use the difference-in-difference identification strategy to estimate the effect of the cut in the legal maximum interest rate using micro-level data from a major credit-rating company. The legal maximum rate cut significantly lowers the loan interest rate and default rate of low-credit borrowers (i.e., high-credit-risk borrowers) in the non-banking sector. However, this effect is limited to borrowers who have not been excluded from the market despite the legal maximum interest rate cut. The loan approval rate of low-credit borrowers decreased significantly after the legal maximum interest rate cut. Meanwhile, the loan approval rate of high-credit and medium-credit (i.e., low credit risk and medium credit risk) borrowers increased. This implies that financial institutions in the non-banking sector should reduce the loan supply to low-credit borrowers who are no longer profitable while increasing the loan supply to high- and medium-credit borrowers.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권8호
• /
• pp.497-502
• /
• 2021

This study aims to investigate the determinants of default risk of rural banks in East Java, Indonesia. The method used is descriptive verification and logistic regression analysis. The data used is secondary in the form of monthly annual financial reports of rural banks in East Java during the period 2009-2018. From the results, it was shown that net interest margin (NIM) as a proxy of market risk, non-performing loan (NPL) as a proxy of credit risk, operation efficiency as a proxy of operational risk and return on assets (ROA) as a proxy of profitability have a significant influence on default risk. Meanwhile, the loan to deposit (LDR) ratio as a proxy of liquidity risk has no significant influence on default risk. Banks need to implement risk management and meet the capital adequacy requirements of regulators so that they are resistant to risk, and also, compliant with bank governance to be able to produce high returns for rural banks have an impact on sustainability and its existence. The ability to identify setbacks in bank conditions and the ability to distinguish between healthy and problematic banks will enable to anticipate default banks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3627-3641
• /
• 2021

Fintech, which stands for financial technology, is growing fast globally since the economic crisis hit the United States in 2008. Fintech companies are striving to secure a competitive advantage over existing financial services by providing efficient financial services utilizing the latest technologies. Fintech companies can be classified into several areas according to their business solutions. Among the Fintech sector, peer-to-peer (P2P) lending companies are leading the domestic Fintech industry. P2P lending is a method of lending funds directly to individuals or businesses without an official financial institution participating as an intermediary in the transaction. The rapid growth of P2P lending companies has now reached a level that threatens secondary financial markets. However, as the growth rate increases, so does the potential risk factor. In addition to government laws to protect and regulate P2P lending, further measures to reduce the risk of P2P lending accidents have yet to keep up with the pace of market growth. Since most P2P lenders do not implement their own credit rating system, they rely on personal credit scores provided by credit rating agencies such as the NICE credit information service in Korea. However, it is hard for P2P lending companies to figure out the intentional loan default of the borrower since most borrowers' credit scores are not excellent. This study analyzed the voices of telephone conversation between the loan consultant and the borrower in order to verify if it is applicable to determine the personal credit score. Experimental results show that the change in pitch frequency and change in voice pitch frequency can be reliably identified, and this difference can be used to predict the loan defaults or use it to determine the underlying default risk. It has also been shown that parameters extracted from sample voice data can be used as a determinant for classifying the level of personal credit ratings.

• KDI Journal of Economic Policy
• /
• 제38권1호
• /
• pp.71-91
• /
• 2016

This paper examines the effect of the government college loan program in Korea on student academic performance, dropout decisions and loan defaults. While fairness in educational opportunities has been guaranteed to some degree through this program, which started in 2009, there has been a great deal of controversy over its effectiveness. Empirical findings suggest that recipients of general student loan (GSL) lower academic performance than those who received income contingent loan (ICL). Moreover, for students attending private universities, a higher number of loans received increased the probability of a dropout decision, and students from middle-income households had a higher probability of being overdue than students from low-income households. These findings indicate that expanding the ICL program within the allowance of the government budget is necessary. Furthermore, providing opportunities for students to find various jobs and introducing a rating system for defaulters are two necessary tasks.

• The Journal of Asian Finance, Economics and Business
• /
• 제5권2호
• /
• pp.35-42
• /
• 2018

In recent decades, financial crises in various countries have often been preceded by the rise in non-performing loans (NPLs) in the banks' asset portfolios. The increase in NPLs is proven to have adverse impact on the banking sector so that understanding the determinant of NPLs is immensely crucial to ensure the efficiency and soundness of the overall economy. This study aims to shed light on bank-specific factors that affect loan default problems in developing countries whose banking sectors play a major role in the overall economy. This study analyzes panel data sets of 36 commercial banks listed in the Indonesian Stock Exchange during the period 2008-2015. Applying fixed-effects panel regression model reveals that Indonesian banks' profitability and credit growth negatively influence the number of NPLs. Moreover, banks with higher profitability are proven to have lower NPLs because they can afford adequate credit management practices. Likewise, banks with higher credit growth evidently have lower NPLs in the sense that they demonstrate more specialized lending activity and thus have better credit management systems. These findings imply that, in order to lower loan defaults that can deteriorate banks' asset quality, banks should maintain their level of profitability and increase, rather than decrease, their credit supply to debtors.

• Journal of Information Processing Systems
• /
• 제17권1호
• /
• pp.75-88
• /
• 2021

Automobile credit business has developed rapidly in recent years, and corresponding default phenomena occur frequently. Credit default will bring great losses to automobile financial institutions. Therefore, the successful prediction of automobile credit default is of great significance. Firstly, the missing values are deleted, then the random forest is used for feature selection, and then the sample data are randomly grouped. Finally, six prediction models of support vector machine (SVM), random forest and k-nearest neighbor (KNN), logistic, decision tree, and artificial neural network (ANN) are constructed. The results show that these six machine learning models can be used to predict the default of automobile credit. Among these six models, the accuracy of decision tree is 0.79, which is the highest, but the comprehensive performance of SVM is the best. And random grouping can improve the efficiency of model operation to a certain extent, especially SVM.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 춘계학술발표대회
• /
• pp.624-626
• /
• 2023

Credit scoring is a technique used by financial institutions to assess the creditworthiness of potential borrowers. This involves evaluating a borrower's credit history to predict the likelihood of defaulting on a loan. This paper presents an ensemble of two Transformer based models within a framework for discriminating the default risk of loan applications in the field of credit scoring. The first model is FinBERT, a pretrained NLP model to analyze sentiment of financial text. The second model is FT-Transformer, a simple adaptation of the Transformer architecture for the tabular domain. Both models are trained on the same underlying data set, with the only difference being the representation of the data. This multi-modal approach allows us to leverage the unique capabilities of each model and potentially uncover insights that may not be apparent when using a single model alone. We compare our model with two famous ensemble-based models, Random Forest and Extreme Gradient Boosting.