Application of artificial neural networks for forecasting corporate bankruptcy Cover Image

Zastosowanie sztucznych sieci neuronowych do prognozowania upadłości przedsiębiorstw
Application of artificial neural networks for forecasting corporate bankruptcy

Author(s): Tomasz Pisula
Subject(s): Economy
Published by: Wydawnictwo Uniwersytetu Ekonomicznego we Wrocławiu
Keywords: bankruptcy prediction; neural networks; logit model; discriminant analysis

Summary/Abstract: In today’s deepening economic crisis there are a lot of companies at risk of bankruptcy. Effective prediction of bankruptcy is one of the most important risk management issues. In recent years, because of greater availability of specialized software packages on the market, the models of artificial intelligence to predict the bankruptcy of companies have been extensively used. The article presents the possibility of using artificial neural networks for the classification of businesses at risk of bankruptcy. On the basis of the research sample of 207 Polish companies which declared bankruptcy in the period from January 2007 to December 2010 there was conducted the sector analysis of bankruptcy as well as there were carried out the empirical studies which compared the efficiency of neural models in relation to the classical parametric models (logit and discriminant analysis).

  • Issue Year: 2012
  • Issue No: 254
  • Page Range: 219-234
  • Page Count: 16
  • Language: Polish