Fuzzy Credit Risk Scoring Rules using FRvarPSO

Abstract

There is consensus that the best way for reducing insolvency situations in financial institutions is through good risk management, which involves a good client selection process. In the market, there are methodologies for credit scoring, each analyzing a large number of microeconomic and/or macroeconomic variables selected mostly depending on the type of credit to be granted. Since these variables are heterogeneous, the review process carried out by credit analysts takes time. The objective of this article is to propose a solution for this problem by applying fuzzy logic to the creation of classification rules for credit granting. To achieve this, linguistic variables were used to help the analyst interpret the information available from the credit officer. The method proposed here combines the use of fuzzy logic with a neural network and a variable population optimization technique to obtain fuzzy classification rules. It was tested with three databases from financial entities in Ecuador — one credit and savings cooperative and two banks that grant various types of credits. To measure its performance, three benchmarks were used: accuracy, number of classification rules generated, and antecedent length. The results obtained indicate that the hybrid model that is proposed performs better than its previous versions due to the addition of fuzzy logic. At the end of the article, our conclusions are discussed and future research lines are suggested.

Associated probabilities aggregations in multistage investment decision-making

“…In the study of Santana et al (2018), the method for credit scoring, analyzing a large number of microeconomic and/or macroeconomic variables selected mostly depending on the type of credit to be granted is developed. This method combines the use of fuzzy logic with a Multistage investment decisionmaking neural network and a variable population optimization technique to obtain fuzzy classification rules.…”

“…Application of the AsF M operators in fuzzy MSDMM for optimal investments Authors often use heuristic approaches in their works dedicated to credit risk assessments. These include fuzzy modeling, fuzzy neural networks, machine learning, genetic programming and more (Breeden, 2021;Chang et al, 2020;De et al, 2019;Deng and Yuan, 2021;Froelich and H ajek, 2019;Hassani et al, 2021;Kamalloo and Saniee Abadeh, 2014;Santana et al, 2018;Shen et al, 2019;Shen et al, 2018;Song et al, 2020;W ojcicka-W ojtowicz and Piasecki, 2021;Wu et al, 2014;Zhang et al, 2015;Zhu et al, 2019).…”

“…, 2020; De et al. , 2019; Deng and Yuan, 2021; Froelich and Hájek, 2019; Hassani et al , 2021; Kamalloo and Saniee Abadeh, 2014; Santana et al. , 2018; Shen et al ., 2019; Shen et al.…”

“…In the study of Santana et al. (2018), the method for credit scoring, analyzing a large number of microeconomic and/or macroeconomic variables selected mostly depending on the type of credit to be granted is developed.…”

Credit scoring by leveraging an ensemble stochastic criterion in a transformed feature space

“…have been combined in order to improve the credit scoring performance. Some representative examples are the work of Santana et al [59], where the authors combine fuzzy logic, neural networks and a variable population optimization technique, to obtain fuzzy classification rules, and another work from the same authors [40], where they define a method able to reduce the number of classification rules involved in the definition of the credit scoring predictive model, reducing the system decision time. Other representative works are that of Carta et al [12], where the authors adopt a two-step feature space transforming method, with the aim to improve the credit scoring performance, or the work of Zhang et al [69], where the authors propose a novel sparse multi-criteria optimization classifier based on one-norm regularization, linear and nonlinear programming, for the credit risk evaluation.…”