Abstract
The relevance of designing, implementing and using scoring systems for credit risk management today is not in doubt. These systems use the characteristics of the client who wants to get a loan, and assesses risk by predicting manners repayment by the borrower. The basis of such systems are usually based on the model of decision making that is based on one of the approaches bayyesivskyy, multiple regression, discriminant analysis, genetic algorithms, classification tree, logistic regression, neural networks, and others. Each approach has its advantages and disadvantages. This article provides an analysis of the existing algorithms scoring systems. The method ROC-analysis, by which is possible to determine the most efficient model of the decision to grant or denial of credit. It is noted that at present, the final decision is made by an expert. However, the same conditions different people make different decisions, because of personal factors that influence the decision making process. Therefore, the paper proposes to assess the borrower based on multiple models and in case of conflicting decisions of each model, to make the final decision based on the constructed model of collective decision.
Highlights
Актуальность создания, внедрения и использования скоринговых систем для управления кредитными рисками сегодня не вызывает сомнения
This article provides an analysis of the existing algorithms scoring systems
It is noted that at present, the final decision is made by an expert
Summary
Актуальність створення, впровадження та використання скорингових систем для управління кредитними ризиками сьогодні не викликає сумніву. В основу таких систем зазвичай покладена модель прийняття рішення, яка побудована на основі одного з підходів: байєсівский, множинна регресія, дискримінантний аналіз, генетичні алгоритми, дерева класифікації, логістичної регресії, нейронні мережі та інші. The relevance of designing, implementing and using scoring systems for credit risk management today is not in doubt These systems use the characteristics of the client who wants to get a loan, and assesses risk by predicting manners repayment by the borrower. The basis of such systems are usually based on the model of decision making that is based on one of the approaches bayyesivskyy, multiple regression, discriminant analysis, genetic algorithms, classification tree, logistic regression, neural networks, and others. Одним із них є оцінка кредитоспроможності позичальника, яка здійснюється за допомогою скорингових систем, що спеціально розробляються для підвищення ефективності прийняття рішень щодо кредитних угод. Середні втрати в підрахунках на одного клієнта становить і є мінімальним при виборі
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Економічний вісник Національного технічного університету України «Київський політехнічний інститут»
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
