No open-access location is known for this record yet.
No open-access location is known for this record yet.
This article is about the importance of valuation of companies’ bankruptcy risk and its methods. Bankruptcies of enterprises are one of the most popular economic appearances that cause a lot of negative effects for companies, employees, other associate firms and institutions, also for the state and society. The perspectives and succession of company’s performance is a concern not just of the company itself, but also of other associate subjects such as investors, shareholders, banks, customers, suppliers and other business partners. One of the most popular ways to measure bankruptcy risk this time is E. Altman method. But there are no researches and conclusions about the relevance of this method for Lithuanian companies. After the application of E. Altman method for 56 Lithuanian joint stock companies the result showed, that it produces meaning error. This method puts too many companies to high or very high bankruptcy risk classes. So a more accurate model for bankruptcy risk prediction is needed. It is important to evaluate as many characteristics of the company as it is possible in order to estimate real probability to crash. Company is a complicated object and it is characterized by a lot of various ratios. Besides, there are no concrete values of each ratio that would be received as good or bad. Each company has its strong and weak characteristics. That‘s why it is necessary to valuate not a single ratio, but the system of ratios in order to get the right estimation of the company and its bankruptcy risk. So the methodology that would let join a lot of variables for this purpose is needed. Cluster analysis is the method that lets evaluate and systematize a lot of variables, reduce them to one ratio. Statistical classification is one of the most important and independent research methods of social–economic occurrences and is widely used in all or almost all spheres of science when doing various statistical researches. Classification lets summarize the information, abridge the data set that is needed for the analysis, by making a choice of one or several classes that are concerned. It is especially important when there is a need to compare the objects, because the objects in the same class are similar. This methodology is suitable for bankruptcy risk estimation as well. The point of the methods described in various books is to group the objects so that the similarity of the object in one class would be the largest and the dissimilarity among classes would be significantly large. Such methodology is not the most suitable for this purpose, so another classification method that we have created is offered. The purport of it is to find the outside points in the objects set that are typical representatives of the classes. The classes are combined depending on these typical representatives. When the classification of Lithuanian joint stock companies was made by this principle, the results showed that the only company “Dirbtinis pluoštas“ in 2000 had a very large probability to go bankrupt and it really crashed in 2001. More than half of the enterprises that had the high bankruptcy risk really crash after 2 or more years. Extreme attention must be paid to these companies that have large probability to go bankrupt in two or more years successively. Four the most important ratios for bankruptcy risk estimation were found by regression analysis. They are return on assets, net profit before taxes / assets, liabilities / assets and gross profit margin. According to the improved bankruptcy risk estimation model, suitable for Lithuanian enterprises, each company for its bankruptcy risk valuation purpose can be measured by four ratios and be compared with the typical representatives of the classes. The company is set to the class, where its dissimilarity with the typical representative is the smallest.
Loading…
Stored only when both papers are in the corpus — 17 citations exist upstream.
Papers that cite the same references — the relational fallback while embeddings are still backfilling.
Imported from OpenAlex (CC0) — nothing here is generated.