Usefulness of multi-criteria taxonomy in corporate bankruptcy prediction – the Polish experience

Jacek Welc , Elżbieta Sobczak

Abstract

One of the essential elements of company’s economic and financial evaluation is ratio analysis. It includes computation and interpretation of multiple bankruptcy risk ratios . When applying accounting ratios in a credit risk evaluation analysts typically either compute and interpret several individual ratios (a univariate approach) or apply some pre - estimated parametric multi - variable econometric tools, such as the Altman model or discriminant function . The problem with the first approach is that it does not take into account the inter - relationships between the individual ratios. The latter approach, in turn, calls for statistical models, whose parameters may not be stable in time and in space. In this paper we empirically examine the usefulness of the alternative approach, that is simple taxonomic ranking, in corporate ban kruptcy risk quantification ( on the basis of data from the Polish stock market ) . We found that this simple non - parametric multivariate method outperforms all other tested approaches (including logit model) in discriminating between bankrupt and healthy firms.
Author Jacek Welc (EMaT / DRE)
Jacek Welc,,
- Department of Regional Economy
, Elżbieta Sobczak (EMaT / DRE)
Elżbieta Sobczak,,
- Department of Regional Economy
Pages1854-1862
Publication size in sheets0.5
Book Loster Tomas, Pavelka Tomas (eds.): The 11th International Days of Statistics and Economics. Conference Proceedings, 2017, MELANDRIUM, ISBN 9788087990124, 1922 p.
Keywords in Englishcredit risk analysis, bankruptcy prediction, taxonomy, ratio analysis
URL https://msed.vse.cz/msed_2017/sbornik/toc.html
Languageen angielski
File
Welc_Sobczak_Usefulness_of_multi-criteria_taxonomy_in.pdf 321,25 KB
Score (nominal)15
ScoreMinisterial score = 15.0, 08-03-2019, BookChapterMatConf
Ministerial score (2013-2016) = 15.0, 08-03-2019, BookChapterMatConf
Publication indicators WoS Citations = 0
Citation count*
Cite
Share Share

Get link to the record


* presented citation count is obtained through Internet information analysis and it is close to the number calculated by the Publish or Perish system.
Back