Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/312961 
Authors: 
Year of Publication: 
2024
Citation: 
[Journal:] Contemporary Economics [ISSN:] 2300-8814 [Volume:] 18 [Issue:] 4 [Year:] 2024 [Pages:] 391-407
Publisher: 
University of Economics and Human Sciences in Warsaw, Warsaw
Abstract: 
This study aims to develop nine different consumer bankruptcy forecasting models with the help of three types of artificial neural networks and to verify the usefulness of new, innovative ratios for implementation in personal finance. A learning sample comprising 200 consumers, and a testing sample of 500 non-bankrupt and 500 bankrupt consumers from Poland are used. The author employed three research approaches to using the entry variables to the models. The unique feature of this study is the proposition of the use of newly developed ratios in household finance similar to the financial ratio analysis that is commonly used in corporate finance. The proposed ratios demonstrated high predictive abilities. The paper answers following questions - (a) Are the three commonly implemented types of neural networks useful in forecasting personal bankruptcy risk?; (b) Which forecasting technique is the most effective not only from the viewpoint of overall effectiveness, but also from the perspective of Type I and II errors?; (c) Which research approach (minimalization versus maximization) guarantees maximum effectiveness?; (d) Are the newly developed types of ratios effective in forecasting personal risk bankruptcy? The research identifies and fulfills three gaps in the literature, and also delivers practical solutions for identifying the level of consumer bankruptcy risk. It provides effective solutions for forecasting the risk in terms of usable models and also delivers highly informative ratios that combine demographic and financial indicators in the twelve ratios. It is one of the first attempts to implement ratio analyses in the usage of household finance worldwide.
Subjects: 
consumer bankruptcy
recurrent neural networks
feedforward multilayer network
self-organizingmaps
financial crisis of households
JEL: 
G17
G51
Persistent Identifier of the first edition: 
Creative Commons License: 
cc-by Logo
Document Type: 
Article

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.