Modelling small-business credit scoring by using logistic regression, neural networks and decision trees

2005 ◽  
Vol 13 (3) ◽  
pp. 133-150 ◽  
Author(s):  
Mirta Bensic ◽  
Natasa Sarlija ◽  
Marijana Zekic-Susac
Keyword(s):  
Neural Networks ◽  
Logistic Regression ◽  
Small Business ◽  
Decision Trees ◽  
Credit Scoring

scholarly journals An implementation of ensemble methods, logistic regression, and neural network for default prediction in Peer-to-Peer lending

2021 ◽  
Vol 39 (1) ◽  
pp. 163-197
Author(s):  
Aneta Dzik-Walczak ◽  
Mateusz Heba
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Logistic Regression ◽  
Regression Model ◽  
Financial Institutions ◽  
Logistic Regression Model ◽  
Credit Scoring ◽  
Peer To Peer ◽  
Ensemble Model ◽  
Peer Lending

Credit scoring has become an important issue because competition among financial institutions is intense and even a small improvement in predictive accuracy can result in significant savings. Financial institutions are looking for optimal strategies using credit scoring models. Therefore, credit scoring tools are extensively studied. As a result, various parametric statistical methods, non-parametric statistical tools and soft computing approaches have been developed to improve the accuracy of credit scoring models. In this paper, different approaches are used to classify customers into those who repay the loan and those who default on a loan. The purpose of this study is to investigate the performance of two credit scoring techniques, the logistic regression model estimated on categorized variables modified with the use of WOE (Weight of Evidence) transformation, and neural networks. We also combine multiple classifiers and test whether ensemble learning has better performance. To evaluate the feasibility and effectiveness of these methods, the analysis is performed on Lending Club data. In addition, we investigate Peer-to-peer lending, also called social lending. From the results, it can be concluded that the logistic regression model can provide better performance than neural networks. The proposed ensemble model (a combination of logistic regression and neural network by averaging the probabilities obtained from both models) has higher AUC, Gini coefficient and Kolmogorov-Smirnov statistics compared to other models. Therefore, we can conclude that the ensemble model allows to successfully reduce the potential risks of losses due to misclassification costs.

scholarly journals A Comparison of Artificial Neural Network and Decision Trees with Logistic Regression as Classification Models for Breast Cancer Survival

2020 ◽  
Vol 5 (6) ◽  
pp. 1170-1190
Author(s):  
Venkateswara Rao Mudunuru ◽  
Leslaw A. Skrzypek
Keyword(s):  
Breast Cancer ◽  
Neural Networks ◽  
Logistic Regression ◽  
Artificial Neural Networks ◽  
Decision Trees ◽  
Cancer Survival ◽  
Breast Cancer Survival ◽  
Classification Models ◽  
Classification Methods ◽  
Artificial Neural

In the field of medicine, several recent studies have shown the value of Artificial Neural Networks, decision trees, logistic regression are playing a major role as the predictor, and classification methods. The research has been expanded to estimate the incidence of breast, lung, liver, ovarian, cervical, bladder and skin cancer. The main aim of this paper is to develop models of logistic regression, Artificial Neural Networks, and Decision trees using the same input and output variables and to compare their success in predicting breast cancer survival in woman. To find the best model for breast cancer survival, the sensitivity and specificity of all these models are measured and evaluated with their respective confidence intervals and the ROC values.

scholarly journals Аналитические модели для решения задачи классификации данных

2021 ◽  
Vol 70 (1) ◽  
pp. 42-46
Author(s):  
О.А. Митина
Keyword(s):  
Data Mining ◽  
Neural Networks ◽  
Risk Management ◽  
Logistic Regression ◽  
Credit Risk ◽  
Decision Trees ◽  
Financial Institutions ◽  
Main Task ◽  
Credit Risk Management ◽  
Target Variable

Credit risk management is the main task of banks and other credit institutions. Untimely partial or complete non-repayment of the loan body, as well as the interest part, within the period established by the agreement and in compliance with all the conditions provided for, is one of the main causes of losses of financial institutions. Data mining technologies contain effective tools for building scoring models – neural networks, decision trees, and logistic regression to predict the value of the target variable that allows you to assess the creditworthiness of the client. The purpose of this article is to show the relevance of the problem of data classification on the example of the financial and credit sphere (credit loan).

Regression model of assessment of customer solvency and banking risks in the process of lending

2019 ◽  
pp. 69-73
Author(s):  
Zoryna Yurynets ◽  
Rostyslav Yurynets ◽  
Nataliya Kunanets ◽  
Ivanna Myshchyshyn
Keyword(s):  
Logistic Regression ◽  
Discriminant Analysis ◽  
Regression Model ◽  
Likelihood Ratio ◽  
Credit Scoring ◽  
Bank Loans ◽  
Management Decisions ◽  
Computer Data ◽  
Proposed Model ◽  
Loan Risk

In the current conditions of economic development, it is important to pay attention to the study of the main types of risks, effective methods of evaluation, monitoring, analysis of banking risks. One of the main approaches to quantitatively assessing the creditworthiness of borrowers is credit scoring. The objective of credit scoring is to optimize management decisions regarding the possibility of providing bank loans. In the article, the scientific and methodological provisions concerning the formation of a regression model for assessing bank risks in the process of granting loans to borrowers has been proposed. The proposed model is based on the use of logistic regression tools, discriminant analysis with the use of expert evaluation. During the formation of a regression model, the relationship between risk factors and probable magnitude of loan risk has been established. In the course of calculations, the coefficient of the individual's solvency has been calculated. Direct computer data preparation, including the calculation of the indicators selected in the process of discriminant analysis, has been carried out in the Excel package environment, followed by their import into the STATISTICA package for analysis in the “Logistic regression” sub-module of the “Nonlinear evaluation” module. The adequacy of the constructed model has been determined using the Macfaden's likelihood ratio index. The calculated value of the Macfaden's likelihood ratio index indicates the adequacy of the constructed model. The ability to issue loans to new clients has been evaluated using a regression model. The conducted calculations show the possibility of granting a loan exclusively to the second and third clients. The offered method allows to conduct assessment of client's solvency and risk prevention at different stages of lending, facilitates the possibility to independently make informed decisions on credit servicing of clients and management of a loan portfolio, optimization of management decisions in banks. In order for a loan-based model to continue to perform its functions, it must be periodically adjusted.

Sign in / Sign up

Export Citation Format

Share Document