Year: 2018
Type: article
Abstract: Finance based on a distance to bankruptcy.As can be expected, the results show that the accuracy decreases.Unfortunately, there still lacks a large number of studies useful in practical decision-making dedicated for data of European small and medium-sized manufacturing companies.Focus on decision rules for determining and concretization of the decision-making process, whether the company is threatened by bankruptcy and should resist the crisis, is also missing.Studies report results which are of a technical nature, regarding the setting of classifi ers, without the emphasis on practical and easy-to-handle conclusions.For this purpose, it is advisable to use a method such as DT; especially because of its clarity and possible connection with methods for the selection of key variables, which can reduce the dimension of data.In our research, we used the DT to develop EU manufacturing companies prediction model and we extended the boundary of literature reviews into the area of classifi cation using the DT model because of the limited studies of fi nancial bankruptcy prediction using the DT model with comparison to SVM and more advanced DT models like RF and ADA boosting (also based on DT classifi ers).The paper is organized as follows: In Section 1, the key points on relevant bankruptcy prediction is presented.Section 2 presents the whole research methodology, data used in experiment, together with metrics used for evaluation of prediction accuracy.Section 3 gives a brief description of the obtained result and shows vizualization of decision rules for the data used.Finally, Section 4 highlights the conclusions and directions for the future research and discusses the obtained results.
Source: E+M Ekonomie a Management
Authors: Václav Klepáč, David Hampel
Institution: Mendel University in Brno
Language: English
Subfield: Accounting
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W2795777555