Search

This paper shows how to improve the measurement of credit scoring by means of factor clustering. The improved measurement applies, in particular, to small and medium enterprises (SMEs) involved in P2P lending. The approach explores the concept of familiarity which relies on the notion that the more familiar/similar things are, the closer they are in terms of functionality or hidden characteristics (latent factors that drive the observed data). The approach uses singular value decomposition to extract the factors underlying the observed financial performance ratios of SMEs. We then cluster the factors using the standard k-mean algorithm. This enables us to segment the heterogeneous population into clusters with more homogeneous characteristics. The result shows that clusters with relatively fewer number of SMEs produce a more parsimonious and interpretable credit scoring model with better default predictive performance. [ABSTRACT FROM AUTHOR]

Application scorecards allow to assess the creditworthiness of loan applicants and decide on acceptance. The accuracy of scorecards is of crucial importance for minimizing bad debt loss and maximizing returns. In this paper, we extend upon prior benchmarking studies that experimentally compare the performance of classification techniques to discriminate between good and bad applications. We evaluate a range of cost-sensitive learning methods in terms of their ability to boost the profitability of scorecards. These methods allow to take into account the variable misclassification costs that are involved in rejecting good loan applications and accepting bad loan applications. An approach is proposed to estimate these misclassification costs, and various approaches to handle missing credit bureau scores are evaluated. The results of a case study involving a Romanian nonbanking financial institution (NBFI) indicate that cost-sensitive learning complements the existing state-of-the-art scorecard of the NBFI. The best performing cost-sensitive models are found to increase profitability across the three business channels, with a single-digit improvement for two of the channels and a double-digit increase for the other one. The result is partly explained by the default rate, which is higher for this latter channel and therefore offers greater potential for improving profitability. [ABSTRACT FROM AUTHOR]

Rural microfinance must contend with the triple challenges of isolation, small‐scale transactions, and risk. These challenges result in information asymmetries and transaction costs that render markets for financial services costly or missing. This paper examines how emerging digital technologies (e.g., mobile money, digital credit scoring, and earth observation) can reshape rural markets for savings, credit, and insurance services, especially in developing countries. Although our synthesis of the literature suggests reason for hope in all three domains, the imperfections of these digital technologies require evaluation and oversight if the resulting rural financial system is to be more efficient and equitable than its predecessor. JEL CLASSIFICATION: D14; F63; R51 Related Content: Digital Transformation of the Agricultural Sector: Pathways, Drivers and Policy Implications Related Content: Who drives the digital revolution in agriculture? A review of supply-side trends, players and challenges Related Content: E-commerce's fast-tracking diffusion and adaptation in developing countries [ABSTRACT FROM AUTHOR]

Recognizing potential defaulters is a crucial problem for financial institutions. Therefore, many credit scoring methods have been proposed in the past to address this issue. However, these methods rarely consider the interaction among customers such as bank transfer and remittance. With rapid growth in the number of customers adopting online banking services, such interaction information plays a significant role in assessing their credit score. In this paper, we propose a novel scalable credit scoring approach called CDGAT (Graph attention network for credit card defaulters) for predicting potential credit card defaulters. In CDGAT, a customer's credit score is calculated based on transaction embedding and neighborhood embedding. To obtain the neighborhood embedding, CDGAT first utilizes the Amount-bias Sampling (AbS) strategy to extract a subgraph for each customer. Next, CDGAT directly aggregates neighbors' features according to their influence weights. The experimental results on the dataset from Industrial and Commercial Bank of China (Macau) Limited (ICBC (Macau)) show that CDGAT significantly outperforms the baseline methods. Furthermore, experimental results reveal that the proposed method is also superior to several state-of-the-art Graph Convolutional Neural Network models in terms of scalability and performance. [ABSTRACT FROM AUTHOR]

Predicting creditworthiness is an important task in the banking industry, as it allows banks to make informed lending decisions and manage risk. In this paper, we investigate the performance of two different deep learning credit scoring models developed on the textual descriptions of customer transactions available from open banking APIs. The first model is a deep learning model trained from scratch, while the second model uses transfer learning with a multilingual BERT model. We evaluate the predictive performance of these models using the area under the receiver operating characteristic curve (AUC) and Brier score. We find that a deep learning model trained from scratch outperforms a BERT transformer model finetuned on the same data. Furthermore, we find that SHAP can be used to explain such models both on a global level and for explaining rejections of actual applications. [ABSTRACT FROM AUTHOR]

Logistic regression is a very popular binary classification technique in many industries, particularly in the financial service industry. It has been used to build credit scorecards, estimate the probability of default or churn, identify the next best product in marketing, and many more applications. The machine learning literature has recently introduced several alternative techniques, such as deep learning neural networks, random forests, and factorisation machines. While neural networks and random forests form part of the practitioner's model-building toolkit, factorisation machines are seldom used. In this paper, we investigate the applicability of factorisation machines to some binary classification problems in banking. To stimulate the practical application of factorisation machines, we implement the fitting routines, based on logit loss and maximum likelihood, on commercially available software that is widely used by banks and other large financial services companies. Logit loss is usually used by the machine learning fraternity while maximum likelihood is popular in statistics. Depending on the coding of the target variable, we will show that these methods yield identical parameter estimates. Often, banks are confronted with predicting events that occur with low probability. To deal with this phenomenon, we introduce weights in the above-mentioned loss functions. The accuracy of our fitting algorithms is then studied by means of a simulation study and compared with logistic regression. The separation and prediction performance of factorisation machines are then compared to logistic regression and random forests by means of three case studies covering a recommender system, credit card fraud, and a credit scoring application. We conclude that logistic factorisation machines are worthy competitors of logistic regression in most applications, but with clear advantages in recommender systems applications where the number of predictors typically outnumbers the number of observations. [ABSTRACT FROM AUTHOR]

Summary: Gender is prohibited from use in decision making in many countries. This does not necessarily benefit females in situations of automated algorithmic decisions, e.g. when a credit scoring model is used as a decision tool for loan granting. By analysing a unique proprietary data set on car loans from a European bank, the paper shows that gender as a variable in a credit scoring model is statistically significant. Its removal does not alter the predictive accuracy of the model, yet the proportions of accepted women/men depend on whether gender is included. The paper explores the association between predictors in the model with gender, to demonstrate the omitted variable bias and how other variables proxy for gender. It points to inconsistencies of the existing regulations in the context of automated decision making. [ABSTRACT FROM AUTHOR]

What variables indicate whether a small or medium enterprise (SME) applying for financing from a development bank is a good loan candidate? Structural equation modeling applied to a set of variables can provide the necessary insights. This paper analyzes 407 SMEs that applied for development loans in Pichincha Province, Ecuador. Rather than specific economic/financial quantitative variables often used in credit scoring models, we employed qualitative variables to study creditworthiness. The structural equation methodology can verify whether a client is creditworthy based on company management, product characteristics, and contextual market aspects. Another key contribution is the finding that an entrepreneur's personal/professional traits are the primary determinant for granting this type of loan. The results have theoretical and practical implications that could enhance the limited empirical research in this field to date. [ABSTRACT FROM AUTHOR]

The k-nearest neighbors (k-NN) algorithm, a fundamental machine learning technique, typically employs the Euclidean distance metric for proximity-based data classification. This research focuses on the feature importance infused k-NN model, an advanced form of k-NN. Diverging from traditional algorithm uniform weighted Euclidean distance, feature importance infused k-NN introduces a specialized distance weighting system. This system emphasizes critical features while reducing the impact of lesser ones, thereby enhancing classification accuracy. Empirical studies indicate a 1.7% average accuracy improvement with proposed model over conventional model, attributed to its effective handling of feature importance in distance calculations. Notably, a significant positive correlation was observed between the disparity in feature importance levels and the model's accuracy, highlighting proposed model's proficiency in handling variables with limited explanatory power. These findings suggest proposed model's potential and open avenues for future research, particularly in refining its feature importance weighting mechanism, broadening dataset applicability, and examining its compatibility with different distance metrics. [ABSTRACT FROM AUTHOR]

Credit scoring models help lenders decide whether to grant or reject credit to applicants. This paper proposes a credit scoring model based on boosted decision trees, a powerful learning technique that aggregates several decision trees to form a classifier given by a weighted majority vote of classifications predicted by individual decision trees. The performance of boosted decision trees is evaluated using two publicly available credit card application datasets. The prediction accuracy of boosted decision trees is benchmarked against two alternative machine learning techniques: the multilayer perceptron and support vector machines. The results show that boosted decision trees are a competitive technique for implementing credit scoring models. [ABSTRACT FROM AUTHOR]

Online consumer lending has recently been growing rapidly, but it faces high credit risk. For this problem, developing powerful credit scoring models has become an effective solution and can be achieved from three aspects: modeling approach, data source, and evaluation measure. This paper proposes a novel model that departs from those in previous studies in threefold. First, a heterogeneous deep forest model that combines deep learning architecture and tree‐based ensemble classifiers is proposed as the modeling approach. Second, a Bayesian‐based macroeconomic variable optimization method is developed to determine the macroeconomic variables and the corresponding lag term, and the selected macroeconomic variables are used as supplementary data source for modeling. Lastly, a series of capital charge error measures is proposed to evaluate credit scoring models from a regulatory perspective. The proposal is evaluated on multiple large datasets under performance measures on predictive accuracy, profitability, and capital charge errors. Frequentist and Bayesian nonparametric significance tests are used to examine the statistical significance of heterogeneous deep forest and benchmarks. Three main conclusions can be reached from the comparison. First, heterogeneous deep forest significantly outperforms the industry benchmarks over all the evaluation measures. Second, the predictive performance is enhanced after incorporating the selected macroeconomic variables and the corresponding lag, and the result remains robust under cross‐validation and forward‐chaining validation. Third, the capital charge errors reflect the model performance from a regulatory perspective and thus lead to different rankings from those when evaluating predictive accuracy and profitability. [ABSTRACT FROM AUTHOR]

This study identifies credit risk sources, credit scoring index classification modes and modelling methods, and constructs a credit scoring system for small and micro businesses (SMBs) with soft information. Through the analysis and comparison of neural network models, this study demonstrates the superiority of the back-propagation neural network (BPNN) models for loan classification prediction. There are three contributions and innovations as follows. (1) Based on the actual demands and default characteristics of SMBs, this study adds the behavioural variables of loan managers to strengthen the role of soft information in credit scoring model. (2) It develops a hybrid analysis and comparison framework based on the BPNN model. It identifies that the BPNN model is more suitable for approving SMB loans, as it can precisely identify the second type of error. (3) Using the precious ledger data of SMB loans from a rural commercial bank in Jiangsu province, China, this study compares the prediction accuracy of the credit scoring model based on BPNN using two-level and five-level loan classifications. Further, it illustrates the applicability of the BPNN model. By connecting the practical operations of credit scoring and quantitative models, this paper supports commercial bank examination and approval work of SMB loans. [ABSTRACT FROM AUTHOR]

It is difficult for micro and small entrepreneurial firms (MSEFs) to access external financing from formal financial institutions because financial institutions cannot obtain sufficient and reliable credit information about MSEFs. With the development of the internet and data collection technologies, more and more data can be accessed from different sources, and logistic regression model often suffers from bad performance. In this paper, we propose a credit scoring model using composite MCP logistic regression and firstly apply the proposed method to predict the probability of default of MSEFs in China. The proposed method can carry out parameter estimation and automatic bi-level selection of variables simultaneously with respect to the grouping structure of variables. Empirical results of MSEFs with complex grouping structure data have shown that the proposed method outperforms forward stepwise logistic regression, MCP logistic regression and group MCP logistic regression. In addition, we find that gross salary, tax paid, bank and injury insurance information of MSEFs are the most important factors to predict the probability of default. [ABSTRACT FROM AUTHOR]

In previous research, the Extended Order Scale (EOS) dedicated to risk assessment was analysed. It was characterised by a Numerical Order Scale (NOS) evaluated by trapezoidal oriented fuzzy numbers (TrOFNs). However, the research showed that EOS with two-stage orientation phases, was too complicated. Therefore, the main aim of our paper is to simplify a Complete Order Scale (COS) to a zero- or one-stage order scale and a hybrid approach. For this purpose, a way to calculate the scoring function is presented. The results show that changes in the COS structure influence the values of a scoring function. Replacing just one linguistic indicator gives different results. Another finding of the research is the method's flexibility that allows an expert to individually choose the most suitable COS. The research proves that the boundary between various linguistic labels cannot be precisely defined. However, knowledge of a formal COS structure allows it to be transformed into a less complex one. [ABSTRACT FROM AUTHOR]

This systematic review aims to provide deep insights on emerging trends in, and the potential of, advanced deep learning techniques, such as machine learning algorithms being partially replaced by deep learning (DL) algorithms for credit scoring owing to the higher accuracy of the latter. This review also seeks to explain the reasons that deep belief networks (DBNs) can achieve higher accuracy than shallower networks, discusses the potential classification capabilities of DL-based classifiers, and bridges DL and explainable credit scoring. The theoretical characteristics of DBNs are also presented along with the reasons for their higher accuracy compared to that of shallower networks. Studies published between 2019 and 2022 were analysed to review and compare the most recent DL techniques that have been found to achieve higher accuracies than ensemble classifiers, their hybrids, rule extraction methods, and rule-based classifiers. The models reviewed in this study were evaluated and compared according to their accuracy and area under the receiver operating characteristic curve for the Australian, German (categorical), German (numerical), Japanese, and Taiwanese datasets, which are commonly used in the credit scoring community. This review paper also explains how tabular datasets are converted into images for the application of a two-dimensional convolutional neural network (CNN) and how "black box" models using local and global rule extraction and rule-based methods are applied in credit scoring. Finally, a new insight on the design of DL-based classifiers for credit scoring datasets is provided, along with a discussion on promising future research directions. [ABSTRACT FROM AUTHOR]

In the practice of credit-risk management, the models for receiver operating characteristic (ROC) curves are helpful in describing the shape of an ROC curve, estimating the discriminatory power of a scorecard, and generating ROC curves without underlying data. The primary purpose of this study is to review the ROC curve models proposed in the literature, primarily in biostatistics, and to fit them to actual credit-scoring ROC data in order to determine which models could be used in credit-risk-management practice. We list several theoretical models for an ROC curve and describe them in the credit-scoring context. The model list includes the binormal, bigamma, bibeta, bilogistic, power, and bifractal curves. The models are then tested against empirical credit-scoring ROC data from publicly available presentations and papers, as well as from European retail lending institutions. Except for the power curve, all the presented models fit the data quite well. However, based on the results and other favourable properties, it is suggested that the binormal curve is the preferred choice for modelling credit-scoring ROC curves. [ABSTRACT FROM AUTHOR]

Forecasting the creditworthiness of customers is a central issue of banking activity. This task requires the analysis of large datasets with many variables, for which machine learning algorithms and feature selection techniques are a crucial tool. Moreover, the percentages of "good" and "bad" customers are typically imbalanced such that over- and undersampling techniques should be employed. In the literature, most investigations tackle these three issues individually. Since there is little evidence about their joint performance, in this paper, we try to fill this gap. We use five machine learning classifiers, and each of them is combined with different feature selection techniques and various data-balancing approaches. According to the empirical analysis of a retail credit bank dataset, we find that the best combination is given by random forests, random forest recursive feature elimination and random oversampling. [ABSTRACT FROM AUTHOR]

Bankruptcy prediction is one of the most important business decision-making problems. Intelligent techniques have been employed in order to develop models capable of predicting business failure cases. The present article provides a systematic literature review of the field. As opposed to previous reviews which concentrate on the classification methods, this study adopts a much broader approach to the bankruptcy prediction problem. The survey is articulated around six major axes which cover all the range of issues related to bankruptcy prediction. These axes are the definition of main research objectives, the employed classification methods, performance metrics issues, the input data and data sets, feature selection and input vectors and finally, the interpretation of the models and the extraction of domain knowledge. The findings and employed methodologies of the collected papers are categorized, presented and assessed according to these axes. The ultimate goal is to detect weaknesses and omissions and to highlight research opportunities. We hope that future researchers will find this survey useful in their attempt to orientate their efforts and to locate interesting topics for further research. [ABSTRACT FROM AUTHOR]

Credit scoring is one of the key problems in financial risk managements. This paper studies the credit scoring problem based on the set-valued identification method, which is used to explain the relation between the individual attribute vectors and classification for the credit worthy and credit worthless lenders. In particular, system parameters are estimated by the set-valued identification algorithm based on a given recognition criteria. In order to illustrate the efficiency of the proposed method, practical experiments are conducted for credit card applicants of Australia and credit card holders from Taiwan, respectively. The empirical results show that the set-valued model has a higher prediction accuracy on both small and large numbers of data set compared with logistic regression model. Furthermore, parameters estimated by the set-valued identification method are more stable, which provide a meaningful and logical explanation for extracting factors that influence the borrowers' credit scorings. [ABSTRACT FROM AUTHOR]

Among the numerous alternatives used in the world of risk balance, it highlights the provision of guarantees in the formalization of credit agreements. The objective of this paper is to compare the achievement of fuzzy sets with that of artificial neural network-based decision trees on credit scoring to predict the recovered value using a sample of 1890 borrowers. Comparing with fuzzy logic, the decision analytic approach can more easily present the outcomes of the analysis. On the other hand, fuzzy logic makes some implicit assumptions that may make it even harder for credit-grantors to follow the logical decision-making process. This paper leads an initial study of collateral as a variable in the calculation of the credit scoring. The study concludes that the two models make modelling of uncertainty in the credit scoring process possible. Although more difficult to implement, fuzzy logic is more accurate for modelling the uncertainty. However, the decision tree model is more favourable to the presentation of the problem. [ABSTRACT FROM AUTHOR]

The decision made by machine learning is mostly based on historical data that is used to train them. It raises the awareness that discrimination in machine learning should be eliminated because it may contain societal bias. The financial industry uses credit scoring as a reference to reflect the customer risk profile. To achieve fairness in the model, this paper tries to: (1) assess bias and (2) improve fairness in machine learning models with three bias mitigation methodologies. This study depicts that there is a trade-off between improving fairness and preserving performance. Implementing post-processing methods, for example, Grid Search performs best. [ABSTRACT FROM AUTHOR]

In the last several years, there has been significant research in applying semi-supervised machine learning models to the reject inference problem. When a financial institution wants to build a model to predict the default of credit applicants, the institution only has a known good/bad outcome loan status for the accepted applicants; this causes an inherent bias in the model. Reject inference is used to infer the good or bad loan status of credit applicants that were rejected by a financial institution. This paper presents a reject inference technique in which a semi-supervised framework is developed using a Naive Bayes model. The framework uses the expectation-maximization (EM) algorithm to incorporate rejected applicants into the parameter estimation of the model using a bootstrapping approach. The proposed method has an advantage over traditional reject inference methods because the rejected applicant data will participate in the estimation of the model parameters, thus avoiding the extrapolation problem. The Naive Bayes model using the EM algorithm is compared to logistic regression and several semi-supervised techniques. [ABSTRACT FROM AUTHOR]

For the emerging peer-to-peer (P2P) lending markets to survive, they need to employ credit-risk management practices such that an investor base is profitable in the long run. Traditionally, credit-risk management relies on credit scoring that predicts loans' probability of default. In this paper, we use a profit scoring approach that is based on modeling the annualized adjusted internal rate of returns of loans. To validate our profit scoring models with traditional credit scoring models, we use data from a European P2P lending market, Bondora, and also a random sample of loans from the Lending Club P2P lending market. We compare the out-of-sample accuracy and profitability of the credit and profit scoring models within several classes of statistical and machine learning models including the following: logistic and linear regression, lasso, ridge, elastic net, random forest, and neural networks. We found that our approach outperforms standard credit scoring models for Lending Club and Bondora loans. More specifically, as opposed to credit scoring models, returns across all loans are 24.0% (Bondora) and 15.5% (Lending Club) higher, whereas accuracy is 6.7% (Bondora) and 3.1% (Lending Club) higher for the proposed profit scoring models. Moreover, our results are not driven by manual selection as profit scoring models suggest investing in more loans. Finally, even if we consider data sampling bias, we found that the set of superior models consists almost exclusively of profit scoring models. Thus, our results contribute to the literature by suggesting a paradigm shift in modeling credit-risk in the P2P market to prefer profit as opposed to credit-risk scoring models. [ABSTRACT FROM AUTHOR]

With the in-depth application of artificial intelligence technology in the financial field, credit scoring models constructed by machine learning algorithms have become mainstream. However, the high-dimensional and complex attribute features of the borrower pose challenges to the predictive competence of the model. This paper proposes a hybrid model with a novel feature selection method and an enhanced voting method for credit scoring. First, a novel feature selection combined method based on a genetic algorithm (FSCM-GA) is proposed, in which different classifiers are used to select features in combination with a genetic algorithm and combine them to generate an optimal feature subset. Furthermore, an enhanced voting method (EVM) is proposed to integrate classifiers, with the aim of improving the classification results in which the prediction probability values are close to the threshold. Finally, the predictive competence of the proposed model was validated on three public datasets and five evaluation metrics (accuracy, AUC, F-score, Log loss and Brier score). The comparative experiment and significance test results confirmed the good performance and robustness of the proposed model. [ABSTRACT FROM AUTHOR]

Credit scoring becomes an important task to evaluate an applicant by a banker. Many tools are available for making initial lending decisions. This paper presents a Hidden Markov Model (HMM) for credit scoring, and uses Baum-Welch method; an iterative procedure approach; for building a set of credit scoring models. We introduce HMM/Baum-Welch model: a tool developed to explore a good accurate model for classification problems. There are two phases in this model: learned an initial model from training data using HMM, and re-estimating HMM parameters by an iterative process using Baum-Welch algorithm. The proposed model is successfully applied to a real credit problem, and the application procedure is illustrated through two data sets: German and Australian. The criteria used to evaluate the performance of different resulting models are the accuracy and AUC (area under the ROC curve). The experiment of this model, shows that, HMM with Baum-Welch approach can improve the pattern classification performance in credit scoring. [ABSTRACT FROM AUTHOR]

Emerging markets contain the vast majority of the world's population. Despite the enormous number of inhabitants, these markets still lack a proper finance infrastructure. One of the main difficulties felt by customers is the access to loans. This limitation arises from the fact that most customers usually lack a verifiable credit history. As such, traditional banks are unable to provide loans. This paper proposes credit scoring modeling based on non-traditional-data, acquired from smartphones, for loan classification processes. We use Logistic Regression (LR) and Support Vector Machine (SVM) models which are the top linear models in traditional banking. Then we compared the transformation of the training datasets creating boolean indicators against the categorization using Weight of Evidence (WoE). Our models surpassed the performance of the manual loan application selection process, improving the approval rate and decreasing the overdue rate. Compared to the baseline, the loans approved by meeting the criteria of the SVM model presented a decreased overdue rate. At the same time, using the score generated by a SVM model we were able to grant more loans. This paper shows that credit scoring can be useful in emerging markets. The non-traditional data can be used to build robust algorithms that can identify good borrowers as in traditional banking. [ABSTRACT FROM AUTHOR]

In credit scoring, survival analysis models have been widely applied to answer the question as to whether and when an applicant would default. In this paper, we propose a novel mixture cure proportional hazards model under competing risks. Most existing mixture cure models either do not consider competing risks or generally assume that a subpopulation of subjects is immune to any risk from all the competing risks. Compared with existing models, the proposed model is more flexible since it assumes that a subpopulation of subjects is immune to a subset of risks instead of being immune to all the risks. To estimate model parameters, we derive the likelihood function of the proposed model, based on which an expectation maximization estimation algorithm is developed. A simulation algorithm is designed to simulate time-to-event observations from the proposed model, and simulation studies are conducted to verify the proposed methodology. A real world example of credit scoring for online customer loans based on the proposed model is demonstrated. [ABSTRACT FROM AUTHOR]

In banking and peer‐to‐peer loan applicant firms, customer credit scores have numerous applications in risk control and precision marketing. Numerous credit scoring techniques act as classification methods. In this paper, the main issue is simultaneous and hybrid utilization of the feature selection (FS) algorithm and ensemble learning classification algorithms with respect to their parameter settings to achieve higher performance in the proposed credit scoring model. As a result, this paper reports a hybrid data mining model of generalized fuzzy soft sets (GFSS) theory‐based ensemble learning classification algorithms based on three stages. The first stage addresses data gathering and preprocessing. The second stage uses the adaptive elastic net‐based FS algorithm to eliminate irrelevant or weakly correlated variables. After the appropriate variables are chosen, they are applied to the proposed ensemble model. In the third stage, GFSS theory is exploited to construct a novel weight assignment mechanism for each individual credit scoring model in the ensemble model according to performance. Several comparisons are conducted to validate the proposed model using real world credit datasets. The experimental results, analysis, and statistical tests prove the ability of the proposed approach to improve classification performance against all of the base classifier, hybrid, and combination methods in terms of average accuracy, area under the curve, H‐measure, and Brier score. [ABSTRACT FROM AUTHOR]

The issue of estimating the probability of default constitutes one of the foundations of risk systems applied in modern banking. The Basel Committee pays a lot of attention to ways of its estimation and validation. This paper discusses statistical methods enabling PD estimations with consideration of the retail character of a credit portfolio. The author refers to the issue of defining default and to the way of calculating the number of days in arrears. This paper presents the results of research studies obtained on the basis of retail credit portfolio. For selected sub-portfolios, the author makes a comparison of the probability of default, which enables the explicit risk assessment. [ABSTRACT FROM AUTHOR]

Credit scoring is extensively used by credit industries and financial institutions for financial decision-making. It is a way to assess the risk associated with an applicant based on historical data. However, the historical data may have large number of redundant and noisy features which could affect performance of credit scoring models. Main focus of this paper is to develop a hybrid credit scoring model by combining the feature selection and multi-layer ensemble classifier framework to improve the prediction performance of credit scoring model. The proposed hybrid credit scoring model uses hybrid binary particle swarm optimization and gravitational search algorithm (BPSOGSA) for feature selection and multi-layer ensemble classifier framework with five heterogeneous classifiers. A novel V-shaped transfer function for BPSOGSA is also designed for effective feature selection, which is used to transform the continuous search space to binary search space. Also, a novel fitness function for BPSOGSA is proposed to calculate the fitness value for each search agent. Further, multi-layer ensemble classifier framework along with a novel aggregation function is designed based on generalized convex function. The proposed hybrid credit scoring model is validated using Australian, German-categorical, German-numerical and Japanese credit scoring datasets. The experimental results on all the datasets demonstrate that the proposed credit scoring model outperforms other methods such as random forest and ensemble frameworks, namely majority voting, layered majority voting, weighted voting and layered weighted voting in terms of accuracy, sensitivity, G-measure and ROC characteristics. [ABSTRACT FROM AUTHOR]

Copyright of Acta Economica is the property of University of Banja Luka, Faculty of Economics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This paper proposes a Bayesian network model to address censoring, class imbalance and real-time implementation issues in credit risk scoring. It shows that the Bayesian network model performs well against competing models (logistic regression model and neural network model) along several dimensions such as accuracy, sensitivity, precision and the receiver characteristic curve. Better performance of the Bayesian network model is particularly salient with class imbalance, higher dimensions and a rejection sample. Furthermore, the Bayesian network model can be scaled efficiently when implemented onto a larger dataset, thus making it amenable for real-time implementation. [ABSTRACT FROM AUTHOR]

Ethical banking, microfinance institutions or certain credit cooperatives, among others, grant socially responsible loans. This paper presents a credit score system for them. The model evaluates social and financial aspects of the borrower. The financial aspects are evaluated under the conventional banking framework, by analysing accounting statements and financial projections. The social aspects try to quantify the loan impact on the achievement of Millennium Development Goals such as employment, education, environment, health or community impact. The social credit score model should incorporate the lender's know-how and should also be coherent with its mission. This is done using Multi-Criteria Decision Making (MCDM). The paper illustrates a real case: a loan application by a social entrepreneur presented to a socially responsible lender. The decision support system not only produces a score, but also reveals strengths and weaknesses of the application. [ABSTRACT FROM AUTHOR]

In this paper, a credit scoring integration model incorporating BRNN is used to study the credit scoring problem in automobile finance. Aiming at the problems of existing credit scoring models constructed with shallow architecture and the unidirectional limitation of RNN itself, this paper introduces a BRNN model that superimposes RNN models in two directions. The potential relationship between each credit feature is mined through logistic regression, extreme gradient boosting tree, and bidirectional recurrent neural network algorithms, and the final prediction output is linked to the customer’s overall credit to improve the prediction accuracy. In this paper, we study the application of a credit scoring model based on the improved BRNN model for an auto finance company. Data preprocessing techniques and feature screening methods are used to improve the BRNN model and construct the credit scoring model for auto finance at Company A. The BRNN model is the basis for Company A’s credit scoring model. Based on the comparison with other models, it is concluded that the automobile finance credit scoring IBRNN model constructed based on the improved BRNN model in this paper has an accuracy of 89.6% in classifying the user finance data of Company A on different datasets, which is a significant improvement compared with the other five models.

Conventional loan avenues generally focus more on the formal sector than the unbanked sector. A peer to-peer (p2p) lending platform built on blockchain can help bridge the gap between potential lenders and borrowers in need of money in a secure and decentralized environment. The Ethereum blockchain allows for the creation of smart contracts to perform actions in the network by setting logic rules and conditions thereby removing the need for middlemen and can be inclusive of the unbanked sector. The p2p platform introduces swarm learning for credit scoring, which is a novel methodology that utilizes smart contracts to train decentralized machine learning models. Each training round happens on the local device with the user data, which then exchanges the training parameters and weights to the machine learning model maintained in the smart contract. This allows for preserving the privacy of the user data by ensuring the data never leaves the device but only the inference does. Upon analyzing the user's behavior, a statistical credit score is assessed for validating the chances of the user to default his/her loan repayment. The performance of the proposed model that has been trained using the swarm learning technique is close to the model that had been trained in a centralized environment while overcoming the drawbacks of federated learning by incorporating blockchain and swarm methodology. [ABSTRACT FROM AUTHOR]

Neural networks (NNs) have successfully been applied to classification problems including credit scoring. The tuning of the structural parameters of the NNs has a direct impact on their accuracy. In this paper, a hybrid approach based on the genetic algorithm (GA) is proposed to adjust the structural parameters of a classifier NN to achieve high accuracy. Two well‐known credit scoring datasets—Australian and German datasets—are used to test the proposed approach. The results indicate that the proposed hybrid approach is able to successfully tune the structural parameters, while the accuracy of classification is enhanced and its complexity dramatically diminished in comparison with other existing approaches. The performance of the proposed algorithm has been investigated through statistical analysis The best‐known solutions achieved by the proposed approach have an accuracy equal to 97.78% and 87.1% for Australian and German datasets, respectively. The results indicate 2.68% and 0.1% improvement in comparison with the best results reported in the literature, respectively. This improvement is important for real cases in which millions of loans are allocated using credit scoring approaches. [ABSTRACT FROM AUTHOR]

Segmentation or clustering is a key concept in the economy. It can be defined as a grouping of interconnected firms or individuals in a particular domain. It is a powerful tool that may be used to form clusters in order to increase business productivity and efficiency. The aim of this paper is to investigate the concept of clusters in the development of credit scoring models. We propose a hybrid method based on clustering and random forest techniques. The purpose is to design an effective credit scoring model and enhance financial bankruptcy prediction. We use clustering to partition data into relatively homogeneous groups of credit applicants where the aim is to cluster creditworthy applicants against non-creditworthy ones. We consider k-means algorithm in the clustering step to segment similar applicants into groups. Then, we apply random forest learning technique on the clustered data. Empirical studies are conducted on six well-known financial datasets of different sizes. The numerical results are promising and show the effectiveness of the proposed technique for applicants segmentation. When clustering is used with classification, the resulting method succeeds in improving highly the classification performance. [ABSTRACT FROM AUTHOR]

In a service based economy, companies strive to continue deriving revenue by creating and nurturing long term relationship with clients. A case in point is Retail Banking, where customer value is of utmost importance. In today's hyper-competitive environment, banks are aggressively leveraging their customer base to engage in revenue driving activities such as cross selling and up-selling. To be successful, it is imperative for banks to embrace the power of analytics to gain insights and appropriately evaluate risks and opportunities-enabling more efficient decision making in the quest to enhance share of the wallet. This paper examines the various applications of analytics in Retail Banking and provides pointers for analytic implementations. The paper is divided into 10 parts. Part 1 is Introduction, Part 2 is about Business Analytics, Part 3 is about Data Mining, Part 4 deals with Literature Review, Part 5 is about Theoretical Framework on Analytics in Retail Banking, Part 6 is Methodology and Discussion on results, Part 7 is Inference and Conclusion. [ABSTRACT FROM AUTHOR]

When building models, it is common to split the whole dataset into a development and a validation sample. In some cases, using random sampling instead of stratified sampling can lead to loss of representativeness of final samples. In such cases, a model built on these data gives different or unexpected results when its performance is measured on the validation sample. In the business area, a lack of representativeness can cause interpretative problems and can have a huge financial impact when a biased model is involved in the credit granting process. The aim of this paper is to examine and understand why representativeness should be checked before the start of modelling. The paper deals with methods of identification of selection bias in time. It recommends using three tests as a common part of the data preparation process. [ABSTRACT FROM AUTHOR]

Using data from a Taiwanese finance company, this paper empirically investigates the value of soft information, information that requires the subjective interpretation by the loan officers who collect it and cannot be credibly transmitted to others, for making small business loans. It finds that the use of soft information significantly improves the power of default prediction models. It also identifies the types of soft information that are helpful for predicting loan defaults. In addition, it shows that borrowers with more favorable soft information enjoy lower interest rates. These results imply that soft information is important for small business lending. [ABSTRACT FROM AUTHOR]