Publication detail

Bankruptcy Prediction: To what Degree does Past Development Count?

KARAS, M. REŽŇÁKOVÁ, M.

Original Title

Bankruptcy Prediction: To what Degree does Past Development Count?

Type

journal article - other

Language

English

Original Abstract

Abstract: In most cases, bankruptcy models are based on financial indicators (so called predictors) that describe a current condition or a certain area of financial health, like profitability, indebtedness and so on. But they do not tell us anything about relevant past development in this area. The main question of the research presented was, how much of the information about the past development could be useful in predicting bankruptcy. The aim of our research is to analyse the partial potential of financial ratios for predicting bankruptcy and to compare their importance with the importance of commonly used indicators. Twenty eight indicators were examined in a sample of construction companies operating in the Czech Republic, as well as their development over the past five periods. The non-parametric Boosted Trees method was used to evaluate the relative importance of predictors. The results show that the indicators describing past development could be a significant predictor of bankruptcy, however their main potential is in possible synergy with the indicators describing the current state, both being of the same area of financial health.

Keywords

construction companies, bankruptcy prediction, financial ratio, dynamic indicators, model accuracy, multi-period transformation, model development

Authors

KARAS, M.; REŽŇÁKOVÁ, M.

Released

16. 12. 2016

Publisher

WSEAS

ISBN

2224-2899

Periodical

WSEAS TRANSACTIONS ON BUSINESS AND ECONOMICS

Year of study

13

Number

1

State

Republic of Austria

Pages from

480

Pages to

488

Pages count

488

URL

http://wseas.org/wseas/cms.action?id=12125

BibTex

@article{BUT130640,
  author="Michal {Karas} and Mária {Režňáková}",
  title="Bankruptcy Prediction: To what Degree does Past Development Count?",
  journal="WSEAS TRANSACTIONS ON BUSINESS AND ECONOMICS",
  year="2016",
  volume="13",
  number="1",
  pages="480--488",
  issn="2224-2899",
  url="http://wseas.org/wseas/cms.action?id=12125"
}