The credit risk is one of the main risks in commercial banks and the ability to manage it meaningly affects banks' stability. This risk arises due to the particular reasons related to the possibility to lose loans if the debtors are not able to meet their financial obligations. When making the decisions of financing the loan applicants, banks use the credit risk assessment models that allow estimating the probability of the potential borrowers to default on their loan commitments. The main goal of managing the credit risk in banks is to compound the loan portfolio of the acceptable risk level. According to Derelioglu and Gurgen (2011) the credit risk analysis aims to decrease future losses by estimating the potential risk and eliminating the new credit proposal if the risk is higher than a defined tolerance value. In this respect, it is essential to identify the main factors causing this risk in order to manage it. When assessing the credit risk of every company, banks usually analyze the financial data and some qualitative factors as the independent variables in the statistical credit risk assessment models. But in changing the credit policy in banks and pricing the credits, it is very important to predict the quality of loan portfolio in future. The problem can be summarized as finding the statistical methods that relates the proportion of doubtful and non-performing credits in the loan portfolio (dependent variable) with the set of explanatory variables (macroeconomic information of a country). The aim of this research is to find the macroeconomic determinants that significantly influence the changes of loan portfolio credit risk in banks and to develop the statistical model for prediction of the proportion of doubtful and non-performing loans. The scientific literature analysis results confirmed the influence of macroeconomic conditions on credit risk of debtors in banks and presented that the changes in quality of loan portfolio in banks depend on GDP, inflation, interest rates, money supply, industrial production index, current account balance and other. In empirical research 22 EU countries were grouped into 3 clusters according to their similarity in changes of the doubtful and non-performing loans percentage in banks. The set of 20 independent variables as factors determining the changes in amount of doubtful and non-performing loans was created. These variables were calculated from 9 macroeconomic indicators of 3 years. The model was developed to classify the countries into clusters applying the logistic regression, factor analysis and probit methods. The classification accuracy is 100 %. The predictions of doubtful and non-performing loans indexes are based on the analysis of the scores of extracted 5 factors as new independent variables. The multiple regression and polynomial regression methods were applied for the index predictions in clusters. The developed model in this research enables to predict the percentage of doubtful and non-performing loans in banks with the average 98,06% accuracy. The res...

This article presents an analysis of macroeconomic factors and their impact on the percentage of nonperforming loans (NPLs) in commercial banks of the EU countries. This problem is relevant because in re cent years many EU countries had the economic downturns that can be visible in the main macroeconomic indicators. Also, banks have met the growth of nonperforming loans when the debtors were not able to meet their financial obligations. The Basel III Agreement notes the necessity to consider the economic conditions of a country when assessing the credit risk of loan applicants. The results of this research can be useful for banks, because the main relations between macroeconomics and nonperforming loans have been revealed. Since 2009, Lithuania has one of the highest NPL percentage in the EU, and the meaningful impact of economic de terioration on the debtors' ability to repay debts to banks has been proven. The same situation was ascertained in other EU countries with imperfect economic conditions. Conversely, it has been estimated that banking sys tems in the EU countries with developed economies are not very sensitive to the business cycle fluctuations. So, in Lithuanian banks, when managing credit risk, the consideration of economic conditions is very important.

The attribution of credit ratings for clients is a very important issue in the banking sector. Banks must evaluate credit risk of credit applicants by using standardized (external rating institutions) or internal ratings-based (IRB) methods. Banks which decided to use IRB method attempt to develop precise internal credit rating models for the evaluation of creditworthiness of their borrowers.The internal rating method for the estimation of default probability requires to collect the default information from the historical data in banks. The major studies about default determinating factors are based on classification methods (Zhou, Xie, Yuan, 2008). A classification model considers the default measurement as the pattern recognition where all borrowers are divided to non-default and default groups based on their financial and non financial data. Banks attempt to construct an evaluation model that can be used to discriminate new sample.This research focuses on a credit rating model development which could attribute credit ratings for Lithuanian companies. The steps of a model's development and improvement process are described in this paper.The model's development begins with the selection of initial variables (financial ratios) characterizing default and non-default companies. 20 financial ratios of 5 years were calculated according to annual financial reports. Then statistical and artificial intelligence methods were selected for the classification of companies into two groups: default and non-default. A discriminant analysis, logistic regression and artificial neural networks (multilayer perceptron) were applied for this purpose. Often statistical methods are not able to operate with a large amount of data, so the analysis of variance, Kolmogorov-Smirnov test and factor analysis were applied for data reduction. Artificial neural networks often are able to analyze a large amount of data so variable selection was accomplished by the network itself calculating ranks of importance for every initial variable. There were constructed 15 classification models and their classification accuracy was measured by calculating correct classification rates. The most accurate was a logistic regression model analyzing data of 3 years (97% of correctly classified companies). Then the sample of companies was supplemented with new data and changes in classification accuracy were estimated. The significant decrease of classification accuracy conditioned the need of model update. For this reason the logistic regression coefficients were recalculated. In order to classify nondefault companies into 7 classes: profitability, liquidity, financial structure and individual possibility of default estimated by a logistic regression model were determined as rating criterions. Then the rating scale was constructed and credit ratings were attributed for companies in the sample. The calculated probabilities of default indicated that some lower ratings have lower probabilities of default. These imperfections were corrected by the modification of a...

The assessment and modeling of the credit risk is one of the most important topics in the field of financial risk management. In this investigation the credit risk assessment model was developed and tested for Lithuanian companies. 20 financial ratios of the companies were calculated for each year of the 3 year period of interest. The analysis of variance (ANOVA) and Kolmogorov-Smirnov test were applied and the set of variables reduced from 60 to 25. Logistic regression was used for the classification of the companies into reliable and not reliable ones. Financial ratios, having the highest correlation to the possibility of default were selected for further investigation and several credit ratings were attributed to the companies according to these variables’ values. The average values of Mahalanobis Distances calculated for the most reliable companies were the lowest and these values increased with a decreased reliability of the company. The differences between Mahalanobis Distances of the companies having different credit ratings confirmed the reliability of the model results. Santrauka Kredito rizikos vertinimas ir modeliavimas – viena iš aktualiausiu temų, kalbant apie finansinės rizikos valdymą. Atlikto tyrimo metu buvo sukurtas kredito rizikos modelis. šis modelis išbandytas 198 Įmonių aibėje, skaičiuojant po 20 finansinių rodiklių 3 analizuojamų metu laikotarpiu. Panaudojus ANOVA metodą ir Kolmogorovo – Smirnovo statistiką, kintamųjų kiekis buvo sumažintas nuo 60 iki 25 rodiklių. Įmonįu klasifikavimui į 2 grupes: patikimus ir nepatikimus banko klientus, atsižvelgiant į jų įsipareigojimų nevykdymo tikimybę, buvo naudojama logistinė regresija. 97 proc. patikimų (nebankrutavusių) ir 82 proc. nepatikimų (bankrutavusių) įmonių suklasifikuotos teisingai. Tolimesniam tyrimui atrinkti 7 finansiniai rodikliai, kurių koreliacinis ryšys su įsipareigojimų nevykdymo tikimybe buvo didžiausias. Atsižvelgiant į šių kintamųjų reikšmės, įmonėms buvo priskirti 9 kredito reitingai. Vidutines Mahalanobio atstumu reikšmes, apskaiČiuotos patikimiausioms kompanijoms buvo mažiausios; šios reikšmės didėjo, mažejantįmonių patikimumui. Skirtingį reitingį įmonėms apskaiČiuoti Mahalanobio atstumų skirtumai, pagrindė modelio rezultatų patikimumą.

This article presents on analysis of macroeconomic conditions in the EU countries in relation loan portfolio to credit risk and banking system interest income. The changing economic environment of banks influences their risks and activity results, so it is important to find the macroeconomic indicators that can determine the changes in debtors’ credit risk and banks’ financial condition. The banking system performs very important functions in a country’s financial system, so for its stability it is important to be able to predict the financial results of the banking system in relation to changes in the economic environment. The new Basel III Agreement seeks to improve the financial sector’s resistance to the possible negative scenarios in the economy and motivates to develop the credit risk assessment models considering their dependence on business cycles. For this reason, the statistical dependence between the set of macroeconomic factors and the loan portfolio credit risk together with interest income were estimated in this research. A statistical classification and regression tree model was developed, which allows to predict the possible changes in the interest income of a country’s banks with the 82.7% accuracy.