The main purpose of this paper is to provide evidence of financial leverage, liquidity, profitability, and firm size ability to predict bankruptcy of Swedish small and medium-sized enterprises (SMEs), and to create a bankruptcy prediction model for Swedish SMEs. The sample consists of 1086 Swedish SMEs, among which 543 did go bankrupt between 2015 and 2019. The paper employs logistic regression and Mann-Whitney U-test to test the hypotheses. The independent variables are derived from previous research and further filtered in a selection process, resulting in a final set of six variables. Financial leverage, liquidity, profitability, and firm size is found to have significantly predictive abilities to determine SME bankruptcy. The model has an overall classification accuracy of 77.6% out-of-sample and is able to classify 82.2% of the bankruptcies correctly out-of-sample.

Yes / Prediction of corporate failure is one of the major activities in auditing firms risks and uncertainties. The design of reliable models to predict bankruptcy is crucial for many decision making processes. Although a large number of models have been designed to predict bankruptcy, the relative performance evaluation of competing prediction models remains an exercise that is unidimensional in nature, which often leads to reporting conflicting results. In this research, we overcome this methodological issue by proposing an orientation-free super-efficiency data envelopment analysis model as a multi-criteria assessment framework. Furthermore, we perform an exhaustive comparative analysis of the most popular bankruptcy modeling frameworks for UK data including our own models. In addition, we address two important research questions; namely, do some modeling frameworks perform better than others by design? and to what extent the choice and/or the design of explanatory variables and their nature affect the performance of modeling frameworks?, and report on our findings.

Tyrimo objektas – įmonių finansinių sunkumų diagnozavimas. Darbo tikslas – sukurti įmonių finansinių sunkumų diagnozavimo modelį ir jį patikrinti pasirinktų įmonių pavyzdžiu. Tyrimo uždaviniai:  išanalizuoti ir susisteminti anksčiau sukurtus bankroto prognozavimo modelius ir jų testavimo rezultatus;  sudaryti įmonių finansinių sunkumų diagnozavimo modelį;  patikrinti sukurto modelio tinkamumą pasirinktų įmonių tarpe. Tyrimo metodai. Analizuojant bankroto prognozavimo modelius ir jų testavimo rezultatus atlikta mokslinės ir metodinės literatūros loginė ir lyginamoji analizė bei sintezė. Įmonių finansinių sunkumų diagnozavimo modeliui parengti skaičiuoti finansiniai santykiniai rodikliai bei atlikta statistinė analizė, rezultatai pateikti taikant monografinį metodą. Remiantis indukcijos metodu ir logine analize tikrintas sukurto finansinių sunkumų diagnozavimo modelio tinkamumas analizuotų įmonių tarpe. Tyrimo rezultatai. Pirmojoje darbo dalyje išanalizuota finansinių sunkumų esmė bei juos sąlygojančios priežastys, nustatytas finansinių sunkumų diagnozavimo būtinumas, išnagrinėti ir apibendrinti bankroto prognozavimo modeliai bei išskirti dažniausiai juose naudojami finansiniai santykiniai rodikliai. Antrojoje darbo dalyje sukurtas logistine analize paremtas finansinių sunkumų diagnozavimo modelis bei iškelta hipotezė, kad jis leidžia patikimai apskaičiuoti įmonių finansinių sunkumų tikimybę tirtų įmonių aibėje. Trečiojoje darbo dalyje sudarytas finansinių sunkumų... [toliau žr. visą tekstą] / Research object – diagnosis of company financial difficulties. Research aim - to create diagnostic model for company financial difficulties and to test it using cases of selected companies. Objectives:  to analyze and systematize previously developed bankruptcy prediction models and their test results;  to create a diagnostic model for company financial difficulties;  to analyze the relevance of the model with the selected companies. Research methods: logical and comparative analysis of scientific and methodical literature, statistical analysis, monographic method, induction method, logical analysis. Research results. The first part analyzes the financial difficulties and their underlying causes, establishes the necessity of the financial stress test model, analyzes and summarizes bankruptcy prediction models, and distinguishes the most frequently used financial ratios. The second part creates a logical analysis based diagnostic model for financial difficulties and presents hypothesis, that this model ensures reliable calculation of probability of company financial difficulties among the tested corporations. The third part uses randomly selected companies to test the created diagnostic model for financial difficulties. The developed model is suitable for the manufacturing companies, because it separates financially well standing companies from the ones that are having difficulties; however, it is not as suitable for the services and sales sector, because the model is only... [to full text]

This thesis present concise but comprehensive overview of most important paper dedicated to prediction of corporate bankruptcy, as well as overview of the theory behind the employed models and crucial indicators for quality assessment and comparison of the estimations. Manually collected data includes financial statement, identification information and especially specifications of management and responsible persons. From this point of view, data collected are of high quality and in Czech Republic relatively unique. Noticeable is also multiple imputation method used, current "state-of-the-art" technique for missing data treatment. Practical part concentrates on models estimation for various data setting, when contrasting models on raw and truncated datasets. By smoothing data, significantly better model can be estimated with superior discriminating power on the same data points. Inclusion of macroeconomic variables as well as even more significant governance indicators according to current stage of research, improved estimated models.

Purpose: To investigate if Altman´s Z´-score model, which calculates financial distress, can be applied on companies established in Sweden and if the financial crisis in 2008 made previously healthy companies go bankrupt. Methodology: Quantitative studies with a positivistic foundation. Empirical data will be collected in order to examine if there is generalizability among the studied objects. Conclusions will be made by comparing the empirical data with the theoretical foundation. Financial distress in firms will be measured. Theoretical perspectives: Altman´s Z´-score model, designed to predict financial distress in private firms. Empirical foundation: A selection of 93 private firms that have gone bankrupt in the years 2008, 2009 or 2010. The firms selected all have a turnover that exceeds 20 million SEK. The years examined will be 2005 to 2009. Conclusion: As this study is carried out, the conclusion is that Altman´s Z´-score model cannot be applied on companies established in Sweden.

Darbo tikslas – sukurti modifikuotą bankroto diagnozavimo modelį ir jį patikrinti Lietuvos įmonių pavyzdžiu. Darbo uždaviniai: 1) ištyrus bankroto veiksnius, nustatyti jo atsiradimo priežastis; 2) atlikti teorinių bankroto tikimybės įvertinimo modelių analizę; 3) atlikus bankroto diagnozavimo modelių teorinę analizę, sukurti ir patikrinti modifikuotą bankroto diagnozavimo modelį. Darbo objektas – bankroto diagnozavimas. Tyrimo metodai: mokslinės literatūros analizė; loginė lyginamoji analizė bei sintezė; dokumentų, turinio analizė, apibendrinimo metodas; statistinė įmonių finansinių rodiklių analizė. Nagrinėjant įvairių autorių mokslinius veikalus apie bankroto diagnozavimo būtinumą, bankroto atsiradimo priežastis, ištirta bankroto tikimybė 6 atsitiktinai pasirinktose įmonėse, remiantis pinigų srautų analizės, E.I.Altmano modeliais bei sukurtas ir išanalizuotas modifikuotas bankroto diagnozavimo modelis. / Object of work – bankruptcy diagnostic. Aim of work – to create modificated bankruptcy diagnostic models and check its use in practise throught Lithuanian companies. Tasks of work: 1) Investigate bankruptcy elements and indentify reasons of bunkraptcy origin 2) To make theoretical analyzes of bankruptcy diagnostic models; 3) To create and develop modificated bankruptcy diagnostic model. The research methods: the analysis of scientific literature, the methods of logistic comparison, the methods of synthesis, content and documents methods, the methods of generalization, statistical financial analyzes. There were analyzed scientific works of various authors about nessecisity to predict bankropt, the reasons of bankruptcy nascency, investigated bankruptcy probability in 6 companies using cash flow, E.I.Altman models and was created and inquired modified bankruptcy model.

Denna uppsats berör ämnet konkurser och behandlar två olika sätt att förutsäga dessa. Dessa tillvägagångssätt är dels revisorers fortlevnadsvarningar, vilket är det tillvägagångssätt som används idag, och dels beräkning med en konkursprediktionsmodell. Syftet med denna studie är att diskutera möjligheten att förbättra träffsäkerheten hos revisorers fortlevnadsvarningar genom att tillämpa en standardiserad fortlevnadsvarning med hjälp av Z”-modellen. Möjligheten undersöks genom att jämföra träffsäkerheten mellan revisorer och Z”-modellen. För att bedöma lämpligheten hos Z”-modellen kartläggs även faktorer och nyckeltal som påverkar revisorers fortlevnadsbedömningar. Studien är av kvantitativ natur och den använda metoden är en dokumentstudie. Studiens urval är samtliga svenska aktiebolag med inledd konkurs under året 2017 och som vid deras senaste bokslut hade en omsättning överstigande tio miljoner kronor samt hade en revisor. Dessa bolag summerar till 336 stycken. De studerade bolagen underkastades en innehållsanalys av årsredovisningarna och tillhörande revisionsberättelser. Från årsredovisningarna inhämtades de siffror som sedan beräknades med hjälp av Z”-modellen, och från revisionsberättelserna inhämtades revisorernas uttalanden om bolagen. Denna data var grunden till hela analysen där träffsäkerheten i att förutsäga en konkurs jämfördes mellan de två olika tillvägagångssätten; revisorernas fortlevnadsvarningar kontra Z”- modellen. Resultatet visar att Z”-modellen är bättre på att förutsäga konkurser än vad revisorer är. Skillnaden i träffsäkerhet analyserades och förklarades med hjälp av teorier som pekar på att revisorerna och deras subjektiva bedömningar kan medföra bias och en underskattning av ett företags negativa siffror. Z”-modellen är å andra sidan objektiv varför dessa problem ej uppstår, vilket verkar medföra en bättre träffsäkerhet. Resultatet visar även att ett statistiskt signifikant samband finns mellan revisorers fortlevnadsvarningar och Z”-modellen. Detta indikerar att revisorer beaktar liknande nyckeltal som Z”-modellen. En faktor som, av studien att döma, påverkar revisorers träffsäkerhet är revisorns tillhörande revisionsbyrå. Detta förklarades av att de olika byråerna använder olika tumregler. / This paper deals with the subject of bankruptcies and deals with two different ways of predicting these. These approaches are partly the auditor's going concern warnings, which is the approach used today, and partly the calculation by a bankruptcy prediction model. The purpose of this study is to discuss the possibility to improve the accuracy of auditors’ going concern warnings by applying a standardized going concern warning with the help of the Z”-model. The possibility is examined by comparing the accuracy in predicting bankruptcies between auditors’ going concern warnings and the Z”-model. Furthermore, to evaluate the suitability of the model, factors and financial ratios that affects the auditors’ judgements are mapped. The method used was of quantitative nature and was a document study. The sample of the study is all Swedish companies that began bankruptcy during the year of 2017 and had a turnover of more than SEK ten million in the last fiscal year and had an auditor. These companies totaled 336. The companies studied were subjected to a content analysis by analyzing the annual reports and associated audit reports. From the annual reports, the figures were then calculated using the Z”- model, and from the audit reports, the auditors' statements about the companies were obtained. This data was the basis for the whole analysis, where the accuracy of predicting bankruptcy was compared between the two different approaches; Z”-model versus auditors' going concern warnings. The result shows that the Z”-model is better in predicting bankruptcy than the auditors. The difference was analyzed and explained using theories that indicate that the auditors and their subjective assessments may lead to bias and an underestimation of a company's negative figures. The Z”-model, on the other hand, is objective why these problems probably do not occur, which ultimately seems to lead to an overall better accuracy. Furthermore, the result shows that a statistically significant relationship exists between the auditors’ going concern warnings and the Z”-model. This indicates that the auditors asses similar financial ratios as the Z”-model. One factor that seems to affect the auditors’ accuracy is the auditor’s audit firm. This was explained by the different firms’ heuristics.

The main aim of the master thesis is to compare of different methodologies of financial health models and bankruptcy prediction models and their cause to company classification. The work deals with the applicability of models on the sample of 45 prosperous companies and 45 companies that were initiating in insolvency process. Sample contain about 33 % companies from building industry, 33 % retail, 16,7 % manufacturing industry and 16,7 % of the other industries mainly services. The special kind of contingency table - the confusion matrix - is used in the methodology to calculate sensitivity, specificity, negative predictive, false positive rate, accuracy, error and other classification statistics. Overall model accuracy is obtained as a difference between accuracy and error. Dependencies of models are acquired based on Pearson´s correlation coefficient. The changes (removing of grey zone and testing new cut-off points) in models are tested in the sensitivity analysis. In practise part there are about 12 financial models calculated (Altman Z´, Altman Z´´, Index IN99, IN01 and IN05, Kralicek Quicktest, Zmijewski model, Taffler model and its modification, Index Creditworthiness, Grunwald Site Index, Doucha´s Analysis). Only two financial indicators (ROA and Sales / Assets) in results were important as crucial part for more than one model. Then are classifications of companies in models determined. It shows that the best models according to overall accuracy are Zmijewski and Altman´s Z´´. On the other hand the worst models are index IN99 and both versions of Taffler´s model. The classification is not caused excessively by extreme values, year of the model creation or country of the origin (hypothesis 1). Based on results it is suggested that the bankruptcy prediction is an accurate forecaster of failure up to three years prior to bankruptcy in most examined models (hypothesis 2). It is observed that the type of model and industry influence the classification of models. In the end, the changes based on sensitivity analysis in the worst companies are made. All of three changes have increased overall classification accuracy of models.

The main objective of our research was to develop a new bankruptcy prediction model for the Czech economy. For that purpose we used the logistic regression and 150,000 financial statements collected for the 2002-2016 period. We defined 41 explanatory variables (25 financial ratios and 16 dummy variables) and used Bayesian model averaging to select the best set of explanatory variables. The resulting model has been estimated for three prediction horizons: one, two, and three years before bankruptcy, so that we could assess the changes in the importance of explanatory variables and models' prediction accuracy. To deal with high skew in our dataset due to small number of bankrupt firms, we applied over- and under- sampling methods on the train sample (80% of data). These methods proved to enhance our classifier's accuracy for all specifications and periods. The accuracy of our models has been evaluated by Receiver operating characteristics curves, Sensitivity-Specificity curves, and Precision-Recall curves. In comparison with models examined on similar data, our model performed very well. In addition, we have selected the most powerful predictors for short- and long-term horizons, which is potentially of high relevance for practice. JEL Classification C11, C51, C53, G33, M21 Keywords Bankruptcy...

The purpose of this doctoral thesis is to create a new bankruptcy prediction model and also to design how to use this model for the purposes of credit risk measuring. The starting-point of this work is the analysis of traditional bankruptcy models. It was found out that the traditional bankruptcy model are not enough effective in the current economic conditions and it is necessary to create a new ones. Based on the identified deficiencies of the traditional models a set of two new model series was created. The first series of the created models is based on the use of parametric methods, and the second one is based on the use of newer nonparametric approach. Moreover, a set of factors which are able to identify an imminent bankruptcy was analyzed. It was found, that significant signs of imminent bankruptcy can be identified even five years before the bankruptcy occurs. Based on these findings a new model was created. This model incorporates variables of static and even dynamic character for bankruptcy prediction purposes. The overall classification accuracy of this model is 92.27% of correctly classified active companies and 95.65% of correctly classified bankrupt companies.