Credit Risk Assessment Research Articles

Improving Credit Risk Assessment through Deep Learning-based Consumer Loan Default Prediction Model

This study aims to enhance credit risk identification, improve loan borrower review efficiency, and increase default prediction accuracy rate using data mining and machine learning techniques. The study also employs deep learning to develop a consumer loan default prediction model that minimizes credit risks and ensures consistent development. The researchers collected data from a survey of 1000 participants, stratified into local and foreign banks, and selected the top 11 banks based on turnover and customer volume. To construct the machine learning model, Keras, a neural network library that runs on TensorFlow, was utilized. The model predicts loan applicant default likelihood. The study's practical implications demonstrate a noteworthy success rate of customer default prediction, which can significantly benefit banks. The model was evaluated on a test set of 250 records and achieved a test set accuracy of 95.2%, correctly predicting the default state of 238 out of 250 respondents.

Credit Risk Analysis Using Fuzzy Logic with Machine Learning Models

Credit Risk is an important issue in the Banking Industry. Credit risk Prediction and assessment of credit is a difficult task for credit managers. Banking Industry has large amount of data related to the behavior of the customer and their credit history, but this raw data is not useful for making correct judgment in credit decisions. The banking industry is need of a correct credit decision making system, to distinguish between good customers and default customers. The data mining domain is suitable for assessing credit risk and making good decisions on credit. Feature selection is one of the main pre-processing step in the data mining. This paper use FuzzyRoughSetTheory (FRST) for finding feature subset. Four other feature selection methods are used to find the optimal feature subset. The four feature selection methods are Information Gain, Relief, Chi-Squared and Wrapper subset model. These different feature selection methods are compared in terms of accuracy and efficiency.

Unleashing the Power of Machine Learning: A Comparative Study of Classification Algorithms for Credit Risk Assessment

Credit risk assessment is a critical task in the financial industry, enabling lenders to evaluate the likelihood of default and make informed lending decisions. This paper presents a comprehensive comparative study of classification algorithms for credit risk assessment using machine learning techniques. The paper commences by providing an overview of the importance of credit risk assessment and the challenges faced by traditional methods. It then delves into the exploration of various machine learning algorithms, including logistic regression, decision trees, random forests, support vector machines, and neural networks, highlighting their potential in credit risk assessment. The objective of this study is to compare the performance of different classification algorithms in credit risk assessment. To achieve this, a dataset comprising historical credit data, including borrower information, financial indicators, and repayment history, is collected. The dataset is preprocessed to handle missing values, outliers, and feature engineering is applied to extract relevant predictors. A comprehensive evaluation is conducted, considering performance metrics such as accuracy, precision, recall, and area under the receiver operating characteristic curve (AUC-ROC). The comparative study provides insights into the strengths and weaknesses of each algorithm and their suitability for credit risk assessment in different scenarios. The results of this study contribute to the existing literature on credit risk assessment and offer practical guidance for financial institutions in selecting appropriate machine learning algorithms. Furthermore, the paper discusses potential challenges and limitations associated with the application of machine learning in credit risk assessment and proposes future research directions.

The effects of non-performing loans on bank stability and economic performance in Zimbabwe

This study explores the impact of non-performing loans (NPLs) on the Zimbabwean banking industry’s stability and economic performance during the dollarization era. The panel vector autoregressive (PVAR) model was applied using annual data from 2009 to 2017. The findings indicated that short-run NPL shocks negatively impact the risk-adjusted return, while the impact on risk-adjusted capitalization is positive but dies off in the long run. The findings from the paper further show that NPLs have a strong negative and significant effect on loan growth and economic performance in the short run but remain muted in the long run. The study results also show a bi-directional causality between banking industry stability and NPLs. In summary, NPLs affect banking industry stability, loan growth and economic performance in Zimbabwe. A possible implication is the formulation of a sound regulatory framework that curbs the increase in NPLs, promotes stability within the banking industry, and improves economic performance. The practical implication is that banks must get it right the first time regarding bank lending policies. Thus, the study recommends that Zimbabwean banks proactively manage their exposure to non-performing loans by implementing rigorous credit risk assessment processes.

Big data analytics implementation issues: A case study of a large bank in Malaysia

This case study discusses the concept of big data analytics, which involves managing and analyzing large, complex data to discover patterns and insights. Implementing big data analytics has many benefits, such as improving decision-making processes, enhancing accounting processes and compliance, boosting marketing campaigns, and providing better credit risk assessment procedures. This case study aims for the student to understand the challenges of big data analytics implementation in a Malaysian commercial bank. Students take on the role of a Senior IT Manager who needs to consider solutions to the implementation problems. Although the team proposed using cloud-based services as a solution to these issues, the CEO expressed concerns about security risks and requested other alternative solutions. The team has also been requested to come up with solutions to the data quality issues experienced by the case study bank.

Credit Risk Assessment and Loan Repayment among Development Financial Institutions. A Case of Kenya Industrial Estates Limited

The informal sector of the Kenyan economy continues to remain the largest employer of a majority of the population. However, it is bedevilled by a variety of challenges, chief of them being inadequate access to credit. To address this, the government has set up development financial institutions such as Kenya Industrial Estates (KIE) and mandated them with the role of availing loans to the sector in order to spur economic growth. The low rate of loan repayment advanced by Development Financial Institutions (DFIs) has necessitated the need to investigate the cause(s) of this adverse trend. This research was undertaken to probe how credit risk assessment affects loan repayment in development financial institutions, with Kenya Industrial Estates Ltd as a case study. The specific objectives were to find out how borrowers’ character and capacity affect loan repayment at KIE. A descriptive design guided the study with primary data being obtained through a questionnaire, whose reliability was tested and found to be acceptable with a Cronbach alpha of 0.795. A census was conducted on all 28 branches in Kenya. Two respondents from each branch were selected purposively to result in a total of 56 respondents. Regression analysis suggested that borrowers’ character and capacity positively and significantly affected loan repayment. The null hypotheses for the two variables were subsequently rejected. Spearman’s correlation coefficient for the two predictor variables further revealed that they influenced loan repayment positively. The study concludes that borrowers’ character and capacity have a profound effect on loan repayment among DFIs. The study recommends that DFIs need screen borrowers’ credit history and debt-to-income ratio thoroughly before advancing loans in order to improve loan repayment.

Credit risk analysis using boosting methods

Abstract The use of credit for various occasions has become a routine in our lives. In return, banking and financial institutions require to determine whether the loan demands from them contain any risk. Accordingly, these institutions have been increased their activities in determining whether credit rating models from past credit records of the person applying for the loan works properly. Machine learning-based technologies have opened a new era in this field. AI and machine learning based methods for credit scoring are currently implemented by banking or non-banking financial institutions. Employed models are to extract meaningful features from the required data in which wide variety of information available. In this study, credit risk assessment is conducted using boosting methods such as CatBoost, XGBoost and Light GBM. To this aim, Kaggle Home Credit Default Risk dataset is used and the effect of crediting tendency on the results is also considered. The results have shown that gradient boosting methods provide results that are close to each other, and crediting tendency produces better AUC score in CatBoost while it causes a small decrement in AUC score of XGBoost and LightGBM.

Carbon Costs and Credit Risk in a Resource-Based Economy: Carbon Cost Impact on the Z-Score of Canadian TSX 260 Companies

Climate risks and climate risk-related policies on carbon threaten the ability of economies to thrive and will impact the credit risk of many sectors, primarily high-emitting sectors. Higher credit risks will also affect lenders if their credit portfolios are exposed to climate change risks. The introduction of carbon pricing policies will exacerbate this threat in resource-based economies such as Canada. While some research exists on climate exposure and risks to lending portfolios, there is a knowledge gap on how carbon pricing impacts individual commercial credit risk. Consequently, this study analyzes the effect of different carbon pricing scenarios on Altman’s z-score. Using the Canadian TSX 260 data between 2010 and 2020 as a sample, this paper applied the costs of different carbon prices using the Canadian Government’s carbon price regime of $0 to $170 to analyze Altman’s z-score variables until 2030. The results suggest that carbon price will significantly impact the credit risk of companies in high-emitting industries, such as the energy sector. We conclude that climate policy exposure in the form of carbon costs will have a real impact on credit risk and that lenders must consider carbon emissions as part of their credit risk assessment.

Stacking ensemble method for personal credit risk assessment in Peer-to-Peer lending

Over the last decade, China’s Peer-to-Peer (P2P) lending industry has been seen as an important credit source but it has recently suffered from a wave of bankruptcies. Using 126,090 P2P loan deals from RenRen Dai, one of the biggest online P2P websites in China, this paper attempts to predict credit default probabilities for P2P lending by implementing machine-learning techniques. More specifically, this study proposes a stacking ensemble machine-learning model to assess credit default risk for P2P lending platforms. A Max-Relevance and Min-Redundancy (MRMR) method is used for feature selection and then irrelevant features are eliminated by using k-means clustering method. Finally, the stacking ensemble model is performed to produce accurate and stable predictions in the feature subset. Experimental results show that stacking ensemble model yields high performance, not only in prediction accuracy but also in precision and recall. In comparison to single classifiers, the stacking ensemble machine-learning model has a minimum error rate and provides more accurate credit default risk prediction. The results also confirm the efficiency of the proposed stacking ensemble model through the area under the ROC curve.

Climate-change scenarios require volatility effects to imply substantial credit losses: shocks drive credit risk not changes in economic trends

IntroductionLong-run Macro-Prudential stability objectives for the banking system have recently motivated a detailed focus on potential future credit risks stemming from climate change. Led by regulators and the NGFS, early approaches apply smooth, top-down scenarios that utilize carbon emissions data combined with physical risk metrics. This general climate stress test approach assesses future credit losses for individual firms and the banking system. While the NGFS approach is in its infancy, a number of discussion points have been raised related to how the approach assesses future credit risks. In contrast to the NGFS approach that focuses on changes to long-run economic growth trends, higher credit risks generally arise from unexpected economic shocks to cashflows and asset values. Systematic shocks that impact many firms like those observed during the last three economic recessions clearly produce higher volatility and systematic deviations from average economic trends.MethodsIn this paper we briefly review aspects of current climate stress test approaches to set the context for our primary focus on assessing future climate induced credit risk and credit risk volatility using a multi credit-factor portfolio framework applied to a benchmark US C&I credit portfolio. First we compare various NGFS climate scenarios using NGFS GDP measures to a CCAR severely adverse stress scenario. We then undertake two additional assessments of future climate driven credit risk by applying an assumed relationship between NGFS global mean temperatures (GMTs) and credit-factor volatilities. All three prospective climate credit risk assessments utilize an empirically-based, credit-factor model estimated from market-based measures of credit risk to highlight the potential role for climate induced increases in volatility. The potential future drivers of volatility could stem from narrower physical risks or broader macro-economic, social or other systematic shocks driven by climate change. All three predicted credit loss assessments suggest that volatility not changes to economic trends ultimately drives higher potential credit risks relating to climate change.ContributionsThe key contributions of this paper are the application of empirically based credit factor models combined with higher climate-driven volatility assumptions that support statistical assessment of how climate change could impact credit risk losses.

Testing Distributions in Banking Sector Loans with Different Computer Programs: An Experimental Analysis for Turkey

Within the fields of risk management and banking, the normality condition is one of the basic assumptions to apply value at risk, capital asset pricing or linear regression models on credit risk assessment. However, banking sector data related to loans may not be normally distributed. Hence, it needs to be put through scientific tests. For this purpose, firstly, Anderson-Darling, Jarque-Bera, Kolmogorov-Smirnov, Shapiro-Wilk, and Shapiro-Francia tests are applied to ninety-two banking sector loan variables and it is demonstrated that most of the variables are not normally distributed. Additionally, the parameters of Normal, Birnbaum-Saunders, Exponential, Extreme Value, Gamma, Generalized Extreme Value, Inverse Gaussian, Log-Logistic, Logistic, Lognormal, Nakagami, Negative Binomial, Non-Parametric, Poisson, Rayleigh, Rician, t- Location-Scale, and Weibull distributions are estimated for loan variables. Thirdly, when the data are not normally distributed, it is necessary to examine the other test results. Therefore, Kolmogorov-Smirnov, Anderson Darling, and Chi-square test results are employed for sixty-one distributions related to the variables and best fitted distribution per variable is aimed at. The results indicate that different computer codes and programs may give different outcomes in connection with the normality and best fitting distribution. Therefore, the use of different strategies may also be adopted in risk management courses along with the traditional ones since the normality assumption is an essential first step for the application of such techniques. Finally, pedagogically speaking, it should be noted that teaching the essence of mathematical background and computer codes could be strategically useful for students in internalizing these distribution concepts. Keywords: distribution, banking sector loans, risk management

Interpretable selective learning in credit risk

Forecasting credit default risk has been an important research field for several decades. Traditionally, logistic regression has been widely recognized as a solution because of its accuracy and interpretability. Although complex machine learning models may improve accuracy over simple logistic regressions, their interpretability has prevented their use in credit risk assessment. We introduce a neural network with a selective option to increase interpretability by distinguishing whether linear models can explain the dataset. Our methods are tested on two datasets: 25,000 samples from the Taiwan payment system collected in October 2005 and 250,000 samples from the 2011 Kaggle competition. We find that, for most of samples, logistic regression will be sufficient, with reasonable accuracy; meanwhile, for some specific data portions, a shallow neural network model leads to much better accuracy without significantly sacrificing interpretability.

Optimasi Parameter Support Vector Machine dengan Algoritma Genetika Untuk Penilaian Resiko Kredit

The aim of this study is to optimize the parameters of a Support Vector Machine (SVM) using a genetic algorithm for credit risk assessment. Consumer credit data from a bank is used in this research. The results show that the SVM with parameters optimized using a genetic algorithm provides better classification performance compared to the SVM with default parameters. In addition, the genetic algorithm can also quickly and efficiently optimize SVM parameters. In conclusion, the genetic algorithm can be used to optimize SVM parameters for credit risk assessment Keywords: Support Vector Machine (SVM), Parameter optimization, Genetic algorithm, Credit risk assessment, Classification performance

Feature Selection Engineering for Credit Risk Assessment in Retail Banking

In classification, feature selection engineering helps in choosing the most relevant data attributes to learn from. It determines the set of features to be rejected, supposing their low contribution in discriminating the labels. The effectiveness of a classifier passes mainly through the set of selected features. In this paper, we identify the best features to learn from in the context of credit risk assessment in the financial industry. Financial institutions concur with the risk of approving the loan request of a customer who may default later, or rejecting the request of a customer who can abide by their debt without default. We propose a feature selection engineering approach to identify the main features to refer to in assessing the risk of a loan request. We use different feature selection methods including univariate feature selection (UFS), recursive feature elimination (RFE), feature importance using decision trees (FIDT), and the information value (IV). We implement two variants of the XGBoost classifier on the open data set provided by the Lending Club platform to evaluate and compare the performance of different feature selection methods. The research shows that the most relevant features are found by the four feature selection techniques.

Credit risk evaluation using clustering based fuzzy classification method

Credit scoring is a crucial indicator for banks to determine the financial position and the eligibility of a client for credit. In order to assign statistical odds or probabilities to predict the risk of nonpayment in relation to many other factors, the scoring criterion becomes an important issue. The focus of this study is to propose a clustering based fuzzy classification (CBFC) method for credit risk assessment. We aim to illustrate the beneficial use of machine learning (ML) methods whose prediction power is increased by adopting fuzzy theory to calculate the default risk with a better selection of the features contributing to it. An important feature of the CBFC method is that membership values obtained as a result of the fuzzy k-means clustering algorithm are used for the purpose of better capturing the structure of an existing system.An extensive comparison is performed to show how CBFC performs compared to the traditional ones for the datasets having different characteristics in terms of the variable types. Five different real-life datasets are studied to expose the contribution of fuzzy approach on improving the ML use. Our findings show that the proposed CBFC models can produce the promising classification results in credit risk evaluation which aid the practitioners and decision makers for issuance of credit purposes.

USE OF ARTIFICIAL INTELLIGENCE IN FINTECH TOOLS IN TERMS OF RISK MANAGEMENT

The financial sector, which is based on knowledge, expertise, and technology, it is a sector that continues its activities by taking advantage of developments and opportunities, at the same time is under the pressure of these developments and is often one of the first implementers of them. As in every technological development, the financial sector is a sector that is forced to transform the most by artificial intelligence algorithms, which provide radical changes and transformations. Artificial intelligence (AI) is increasingly being used in the field of fintech risk management to help financial institutions identify, assess, and mitigate financial risks such as Fraud Detection, Credit Risk Assessment, Operational Risk Management, and Market Risk Management. Overall, the use of AI in fintech risk management has the potential to revolutionize the way that financial institutions manage risk by providing more accurate and sophisticated risk assessments, reducing the time required to identify and mitigate risks, and improving the efficiency of the risk management process. In this study, it is aimed to determine the current situation and make inferences for the future by examining the studies on the use of artificial intelligence in the field of fintech risk management. For this purpose, a literature review was conducted on artificial intelligence algorithms used in fintech tools in terms of risk management. It is thought that the study will contribute to the literature in terms of guiding future research.

Application of Business Intelligence in Decision Making for Credit Card Approval

This paper aims to show how business intelligence can be applied in the credit card approval process. More specifically, the paper investigates how information like an applicant’s age, credit score, debt, income, and prior default can be used in credit card approval prediction.The dataset used for analysis is a publicly available dataset from the UCI machine learning repository. Logistic regression is used to make a prediction model with a reasonable number of attributes for a comprehensible business model. The Chi-square test of independence is used to test the dependence of credit card approval results with attributes. Research uncovers that prior default is supposed to be the most important attribute in the approval process. Finally, the authors propose several visualizations that could help make smarter decisions with effective credit risk assessment.

Credit Risk Assessment in P2P Lending Using LightGBM and Particle Swarm Optimization

The credit risk evaluation is a vital task in the P2P Lending platform. An effective credit risk assessment method in a P2P lending platform can significantly influence investors' decisions. The machine learning algorithm that can be used to evaluate credit risk as LightGBM, however, the results in evaluating P2P lending need to be improved. The aim of this research is to improve the accuracy of the LightGBM algorithm by combining the Particle Swarm Optimization (PSO) algorithm. The novelty developed in this research is combining LightGBM with PSO for large data from the Lending Club Dataset which can be accessed on Kaggle.com. The highest accuracy also presented satisfactory results with 98.094% of accuracy, 90.514% of Recall, and 97.754% of NPV respectively. The combination of LightGBM and PSO shows better results.

Использование марковских моделей со множеством состояний для прогнозирования вероятности дефолта заемщиков

Introduction. After the crises, lenders realized the importance of assessing the risk of default on loan portfolios in various economic conditions. Modeling of credit risk assessment occurs mainly using internal ratings of banks based on probabilistic models of defaults of borrowers over a certain period of time. Theoretical models. Three models are considered. The first is a naive Markov model with R states. The transition matrix is given. The second is a Markov model with multiple states with covariates. Macroeconomic indicators are proposed as covariates. The third model is multinomial logit regression. Approbation of Markov models and multinomial regression on simulated and real data of borrowers’ defaults. We investigate the possibility of using multi-state Markov models to predict borrower defaults in financial institutions over time. Three approaches are considered for credit risk modeling. The first approach assumes that the transition probability matrix is constant over time, and the residuals of the Markov model and logistic regression are taken into account further when forecasting over the time horizon. The second one is supplemented by the Markov model, which takes into account the impact of default risks on migration, both individual factors of borrowers and the economic situation in the country. Using covariates, the models made it possible to simultaneously estimate the transition rate and the probability of erroneous classification of states. A multinomial logistic regression model is considered to compare the results obtained using multi-state Markov models. The proposed models are tested both on real and simulated data. Conclusion. The presented models show good predictive results with high accuracy of default estimates. The models reproduce the structure of the generated data quite well. The peculiarity of the multinomial regression model in predicting defaults is its adjustability, and Markov models estimate the probabilities of defaults. To implement the model, software R was used.

Design and Implementation of an Enterprise Credit Risk Assessment Model Based on Improved Fuzzy Neural Network

ABSTRACT In order to improve the effect of enterprise credit risk assessment, this paper combines the improved fuzzy neural network to construct the enterprise credit risk assessment model. The various jamming patterns that can threaten the credit data transmission system include various blocking jamming and tracking jamming. Moreover, this paper analyzes the bit error performance of the credit data transmission system against various interferences, and obtains the bit error curves against various interferences through program simulation. In addition, this paper builds a variable-speed credit data transmission system based on Simulink, analyzes the key technologies in variable-speed credit data transmission, and simulates the anti-interference performance of variable-speed credit data transmission technology. The simulation study verifies that the enterprise credit risk assessment model proposed in this paper has good risk assessment effect and risk response strategy effect.