Multi-factor fuzzy sets decision system forecasting consumer insolvency risk

Tomasz Korol

doi:10.1007/s40622-024-00399-8

Multi-factor fuzzy sets decision system forecasting consumer insolvency risk

Abstrakt

The objective of this study is to develop a multi-factor decision system predicting insolvency risk for natural persons with the use of fuzzy sets. Considering that the financial situation of households is affected by various endogenous and exogenous factors, the main assumption of this study is that the system for predicting financial difficulties should not be limited to the use of only a few financial variables concerning consumers, but also include variables describing the environment. The author proposes a system consisting of three different forecasting models that connect the macroeconomic and microeconomic environments. It monitors the economic situation of households by also identifying those environmental variables, which may directly, or indirectly, endanger the consumer, such as unemployment rate (job market situation), inflation and interest rates, exchange rates, or economic situation in the country (GDP growth rate, the dynamics of retail sales, etc.). Moreover, the created multi-factor tool is in the form of a flexible application that can be easily adapted to changing economic conditions. Another unique feature of the study is the proposed use of newly developed ratios in household finance, similar to that in financial ratio analysis, which is commonly used in corporate finance. The proposed ratios demonstrated high predictive abilities. The paper also identifies the predictive capabilities of selected macroeconomic variables from the perspective of their impact on the risk of consumer insolvency. The research relies on four samples consisting of a total of 2,400 consumers from Taiwan and Poland. The author created three forecasting models separately for the South-East Asian and Central European regions, and two multi-factor systems, each consisting of 1260 decision rules. The findings clearly showed that a multi-factor system is a significantly more effective method compared to single forecasting models.