The dissertation compares the export credit rating model of the national Export Guarantee and Insurance Agency EGAP with models applied by selected Czech banks. The first part of the dissertation presents a summary of credit risk theory. It depicts the main principles of lending and its risks. The dissertation further describes the factors that influence credit risk and the methods of its modelling. While mathematical risk models project the expected loss as well as its sensitivity to the risk factors, the focus of this thesis lies in qualitative models which set a normalized scale for probability of default, the so called credit rating models. The main contribution of the dissertation lies in the survey carried out among four Czech banks belonging to owners from various countries, from which we get an overview of their rating models. It follows from the gathered information that their models are based on financial indicators when rating the buyers/exporters. The models are also considerably amended by non-financial factors whose importance in certain cases rose following the recent financial crisis. The agency EGAP insures business activities abroad and therefore its model takes into account also specific factors related to the destination country. The main difference between the models of EGAP and the examined banks lies in the method of creation and validation: EGAP does not dispose of sufficient amount of business case studies, so that it has to rely on external consulting services when setting up and validating the model. The dissertation concludes that while all rating models are composed of similar risk factors highlighting past financial indicators of the financed business, each analysed rating model differs significantly in the specific database of business cases that were used to construct the model, depending on the availability of data to the bank/insurer. The conclusion that can be drawn from this fact is that the main factor for successful prevention of future failures of the credit rating models will be the extent of the credit assessment database which will be used for the construction of the respective rating model.

The subject of this thesis is an analysis of causes and impacts of credit risk transfer through credit derivatives on the world economy. It deals with the theoretical view of credit derivatives and their definition. A significant part of the work is devoted to a description of the current credit derivatives market and to motives for trading them. In the continuous process of derivative products innovation selected types are described. Those types that were spread during last two decades. In another part are introduced credit derivatives market participants and special attention is given to the banking sector. Important part is the description of an economic situation in countries in crisis. The aim of this work is to give a complete description of all sectors that were influenced by credit derivatives market.

The priority of this work is to show the importance of risk management and the proper operation in electricity trading companies. In the theoretical part of the thesis are summarized basic information about the energy market in the Czech Republic and also a brief overview of the risks observed in the electricity trading. In recent years there have been several cases where a company did not observe their risks or even did not monitor these risks at all. This had a negative impact on the entire Czech energy market. In the practical part of this thesis are described the cases of companies Moravia Energo a.s. and Korlea Invest a.s., setting its risk management and also consequences of their falls. Furthermore, this work describes the different tools to support risk management, which are not the cheapest definitely. Also work contains an example of calculating the limits for credit exposure of the two companies that I calculated based on the analysis of the annual reports of these companies and I managed to get a lot of interesting information during writing this work that the reader will definitely appreciate.

The diploma thesis is focused on predictive modeling in credit risk management. Banks and financial institutions are mainly interested in it to estimate the probability of client's default in order to make a decision about which client will be accepted and which client will be rejected. The theoretical part includes an introduction of credit scoring and a description of discrete choice models. The linear probability model, the probit model and the logit model are described in detail. The logit model is afterwards used for the prediction of client's default. The practical part is focused on a statistical description of the dataset and a description of how to work with it before we start with the development of the credit scoring model. After that follows the estimation of the model on testing sample, its testing and the estimation of the model on full sample with a description of individual steps of calculation and outputs of the program SPSS.

Diplomová práce se zabývá aplikací fuzzy logiky na proces automatické detekce úpadkového klienta z pohledu řízení úvěrového rizika banky. Na základě analýzy stávajícího informačního systému Credit Risk Monitoring autor navrhuje změnu přístupu v hodnocení úvěrového klienta.

Cette thèse étudie le secteur bancaire tunisien et pourquoi ce dernier se caractérise par des volumesimportants de prêts non performants. Sur un échantillon de 11 banques commerciales cotées durant lapériode 2009-2011, nous examinons l’impact de la structure de propriété et des caractéristiques duconseil d’administration sur le risque de crédit. Nos résultats des régressions sur données de panelrévèlent que ces mécanismes de gouvernance sont défaillants jusqu’à présent et qu’ils ont contribué àl’aggravation des prêts improductifs. A partir de l’apparition de nouvelles tentatives à expliquer lesdéfaillances bancaires par l’apport de la finance comportementale, nous concluons le rôle de l’excès deconfiance dans la gestion de risque de crédit imprudente des banques tunisiennes.Dans le but d’approfondir notre recherche, il nous semble intéressant de vérifier l’impact de l’excès deconfiance sur le comportement de prise de risque d’un responsable de crédits. Les régressionslogistiques multinomiales montrent que l’excès de confiance chez les responsables de crédit évolueavec l’expérience et influence négativement leurs prises de risque de crédit. / This thesis studies the Tunisian banking sector and why it is characterized by large volumes of nonperformantloans. Based on a sample of 11 commercial listed banks during 2009-2011, we examinethe impact of ownership structure and board characteristics on credit risk. Our results reveal thedeficiency of these governance mechanisms.From the appearance of new attempts to explain bank failures by the contribution of behavioralfinance, we try to identify the role of overconfidence in the reckless credit policy. Our results showthat overconfidence recent worsening Non Performants Loans of Tunisian banks.In order to deepen our research, it seems interesting to check the impact of overconfidence on bankers’risk-taking behavior. We extend our work with an experimental study to detect the impact ofoverconfidence on the banker’s risk-taking behavior. Our results of multinomial logistic regressionsshow that banker’s overconfidence evolves with experience and influences negatively his credit risktakingbehavior.

Machine learning is becoming a part of everyday life and has an indisputable impact across large array of industries. In the financial industry, this impact lies particularly in predictive modelling. The goal of this thesis is to describe the basic principles of artificial intelligence and its subset, machine learning. The most widely used machine learning techniques are outlined both in a theoretical and a practical way. As a result, four models were assembled within the thesis. Results and limitations of each model were discussed and these models were also mutually compared based on their individual per- formance. The evaluation was executed on a real world dataset, provided by Home Credit company. Final performance of machine learning methods, measured by the KS and GINI metrics, was either very comparable or even worse than the performance of a traditional logistic regression. Still, the problem may lie in an insu cient dataset, in the improper data prepara- tion, or in inappropriately used algorithms, not necessarily in the models themselves.

The real estate industry is a sector where the companies generally have a capital structure which is high leveraged. The financing – with the related terms – is therefore specifically of high importance for the companies in the sector. Traditionally, the way of obtain financing is by borrowing from the bank. Lately, due to new bank regulations, the banks have become more restrictive in their lending which have lead to a growth of other financing alternatives. For instance, the corporate bond market has grown rapidly. The development has increased the number who acts as lenders. Institutional investors are for an example an actor which invests in corporate bonds. Furthermore, the credit rating agencies plays indirectly an important role in the financing process since their credit ratings are a part in the process of determining the terms. The terms (such as the interest rate) of the financing are mainly based on the credit risk of the company. Since the topic is of big importance and the financing for real estate companies is changing, the main focus is to create further knowledge and understanding regarding the assessments of the credit risk by each actor.  The present thesis shows the credit risk assessment process by each actor where the banks and the credit rating agencies have the most clear framework. The banks and the agencies do a deep assessment which then is discussed in "committees" internally to reach the final assessment.  The investor’s combines own analyses with evaluating earlier credit analysis done by a credit rating agency or a financial advisor in a corporate bond issue.

Since the implementation of the Basel II framework in 2007, banks have been given the  opportunity to apply for the option to develop intern models for calculating their required capital. The purpose with this opportunity is that the capital requirements will correspond to the real risk exposure. This has been criticized, since there are incentives for the bank to do  an incorrect risk assessment intentionally and through that get a lower capital requirement. In this report we study how this opportunity affects the banks’ capitalization and if stricter capital requirements in fact leads to that the Swedish banks are better prepared for a financial crisis. The report also describes the risks that this opportunity to internal rating  causes. The study has been done by qualitative method where seven people, with different interests in the market, have been interviewed. By the answers given by the respondents and by earlier publications this report reveals that stronger capitalization is positive, but that  the Basel framework causes a risk that the banks intentionally underestimates  their risks. Nor is it possible to conclude that the banks are better prepared for a crisis afterthe implementation. This is because the IRB approach is something new and therefore  not optimized and yet balanced.

The banking industry in India has undergone many phases in its history; evolving from a regulated, decentralised system in the early 1800’s, to a regulated, centralised system during British rule, to a nationalised system following India’s independence, and finally a combination of a nationalised and private system adopting global standards as it currently stands. This study has two main aims. Firstly, it will assess the relationship between the firm-specific determinants of capital structure, based on the prevailing literature, and the capital structure of public and private sector banks in India. Secondly, it will determine whether there is a difference in the firm-specific factors that contribute to the determination of the capital structure of public sector banks and private sector banks. This study adopts quantitative methods, similar to previous studies on the relationship between capital structure and its firm-specific determinants. The dependent variable, being total leverage, is regressed against multiple independent variables, being profitability, growth, firm size and credit risk (hereinafter referred to as “risk” unless otherwise indicated) in a multivariate linear regression model. This study adds to the current literature by applying the same firm-specific independent variables to the case of private and public sector banks and then to evaluate and compare the similarities and differences between the regression outputs. The results show that for private sector banks, all independent variables are statistically significant in explaining total leverage, where all the independent variables conform to the current literature on capital structure – profitability (-), firm size (-), growth (+) and credit risk (-). Conversely, for public sector banks, all independent variables were considered to be statistically significant, except for credit risk – profitability (-), firm size (+) and growth (+). These results imply that credit risk is not an important determination in a nationalised banks’ capital structure; thus, providing evidence for the moral hazard theory of public sector banks.