Publication detail

Identifying Bankruptcy Prediction Factors in Various Environments: A Contribution to the Discussion on the Transferability of Bankruptcy Models

REŽŇÁKOVÁ, M. KARAS, M.

Original Title

Identifying Bankruptcy Prediction Factors in Various Environments: A Contribution to the Discussion on the Transferability of Bankruptcy Models

Type

journal article in Scopus

Language

English

Original Abstract

As has been demonstrated by Beaver and subsequently Altman, financial indicators can pick up the risk of impending bankruptcy. This idea led to the construction of bankruptcy models that proved capable of identifying companies threatened with insolvency with great accuracy. A number of authors have demonstrated that the accuracy of bankruptcy models falls significantly if the given model is used in an environment other than that for which it was originally developed. The aim of this article is to identify the financial indicators that are statistically significant predictors of bankruptcy in various environments. The sample investigated is comprised of data on industrial concerns in the Visegrad Four countries for the years 2007 to 2012. A bankruptcy model based on the same set of variables was derived for each country by the method of Boosted Trees. The variables that are statistically significant in all countries and the variables that are specific for individual countries were identified by means of comparison of the significance of the variables in the models created (i.e. in different environments). Most important indicators of bankruptcy prediction can be described as indicators of company size, in our research the value of sales and total assets. Additional significant predictors are debt ratios, liquidity and profitability. However their significance for bankruptcy prediction is different, which is demonstrated by a high degree of variability of these indicators in the surveyed data sample.

Keywords

bankruptcy prediction models, financial ratios, the method of boosted trees, accuracy of bankruptcy models in various environments

Authors

REŽŇÁKOVÁ, M.; KARAS, M.

RIV year

2014

Released

12. 6. 2014

ISBN

1998-0140

Periodical

INTERNATIONAL JOURNAL of MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

Year of study

8

Number

1

State

United States of America

Pages from

69

Pages to

74

Pages count

6

URL

http://www.naun.org/main/NAUN/ijmmas/2014/a042001-065.pdf

BibTex

@article{BUT107878,
  author="Mária {Režňáková} and Michal {Karas}",
  title="Identifying Bankruptcy Prediction Factors in Various Environments: A Contribution to the Discussion on the Transferability of Bankruptcy Models",
  journal="INTERNATIONAL JOURNAL of MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES",
  year="2014",
  volume="8",
  number="1",
  pages="69--74",
  issn="1998-0140",
  url="http://www.naun.org/main/NAUN/ijmmas/2014/a042001-065.pdf"
}