Academic journal article Economics & Sociology

Performance Comparison of Multiple Discriminant Analysis and Logit Models in Bankruptcy Prediction

Academic journal article Economics & Sociology

Performance Comparison of Multiple Discriminant Analysis and Logit Models in Bankruptcy Prediction

Article excerpt

(ProQuest: ... denotes formulae omitted.)


The outset of the global financial crisis is considered to be the year 2008, when the subprime crisis bubble in the United States bursted. Since then, we have observed to a much greater extent business cycle fluctuations resulting in numerous macroeconomic and microeconomic imbalances. Since companies are thought to be the economic subjects operating within the macroeconomic area, they are also affected by the abovementioned imbalances.

Because of the interconnectedness across national economies, Slovak companies also have been experiencing financial difficulties. Often, companies encounter a problem of unpaid bills, secondary insolvency, low law inforcement etc. These deficiences are easily transferred onto other companies. Such a tendency is also reffered to as contagion or knockon effect. Likewise, it should be noted here that bankruptcy law in Slovakia often prefers the debtors' interests over those of creditors. A clear evidence is provided by the well-known instances as Váhostav or Doprastav. Occasionally, there can be a situation when financially distressed company receive investment aid from the state or governmnet bodies. To avoid this type of situation, it is reasonable to have some early warning system capable of timely prediction of such situations.

Over the decades starting from the 1960s up to this point, there have been developed several early warning systems aiming at timely prediction of impeding companies' financial difficulties. It is possible to group them into various categories. In the study by Taffler and Agarwal (2008), these models are divided into two groups: (i) accounting-based models; (ii) market-based models. An application of market-based models in Slovakia is restricted by the underdeveloped capital market. Therefore, in the process of bankruptcy prediction model development, application of accounting-based models is suggested. The use of accounting prediction methods has certain drawbacks, as indicated by the abovementioned authors, including the following: (i) accounting statements present past performance of a firm and may or may not be informative in predicting the future, (ii) conservatism and historical cost accounting mean that true asset values may be very different from the recorded book values, (iii) accounting numbers are subject to manipulation by management, and in addition, (iv) accounting statements are prepared on a going-concern basis. In spite of these limitations, it was confirmed that accounting-based models are not inferior in comparison to market-based models.

Even though application of bankruptcy prediction models is widely spread in Western advanced economies, it has become the fast growing research area in case of transitional economies, including countries of Central and Eastern Europe. Such a growing interest may result from several considerations, e.g. (i) financial institutions under Basel II guidelines are allowed to use their own internal ratings to assess the risk parameters of loan applicants; (ii) joining the EU structures, companies may take the opportunity to receive subsidy. In attempting to distunguish between well-established companies from distressed ones, failure prediction models are used; (iii) expanded activities of private equity financial groups. Specifically, its objective is to find potential company to invest in, merge or acquire. Prediction models may also serve as additonal tools in investors' decision-making.

The focus in this study is on bankruptcy prediction model development based on two various statistical methods as applied to Slovakia. These includes both logistic regression function as well as multiple discriminant function under which we may find financial ratios best distinguishing among healthy and unhealthy companies. Thus, the main objective of this study is to compare the performance of the two proposed bankruptcy prediction models on a sample of selected firms operating in Slovak economic environment. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.