This thesis deals with credit risk and selected methods of its evalua- tion. It is focused on assumptions, calculation methods, results and specifics of the CreditMetrics and the CreditRisk+ models. The CreditRisk+ model analytically determines the portfolio credit losses distribution that is caused by defaults of counterparties. In the CreditMetrics model, the credit migration risk is addition- ally considered and the future portfolio value distribution is calculated using the Monte Carlo simulation. The third approach covered in this thesis is the Solvency II, the set of requirements proposed by the European Union for determination of regulatory capital for insurance companies. In the practical part the three ap- proaches are applied on a set of three portfolios of different credit quality. Their results, particularly the determined level of capital required to cover the risk of unexpected credit losses, are analyzed and compared.

This thesis deals with credit risk and selected methods of its evalua- tion. It is focused on assumptions, calculation methods, results and specifics of the CreditMetrics and the CreditRisk+ models. The CreditRisk+ model analytically determines the portfolio credit losses distribution that is caused by defaults of counterparties. In the CreditMetrics model, the credit migration risk is addition- ally considered and the future portfolio value distribution is calculated using the Monte Carlo simulation. The third approach covered in this thesis is the Solvency II, the set of requirements proposed by the European Union for determination of regulatory capital for insurance companies. In the practical part the three ap- proaches are applied on a set of three portfolios of different credit quality. Their results, particularly the determined level of capital required to cover the risk of unexpected credit losses, are analyzed and compared.

Thesis Portfolio Credit Risk Modeling focuses on state-of-the-art credit models largely implemented by banks into their banking risk-assessment and complementary valuation system frameworks. Reader is provided in general with both theoretical and applied (practical) approaches that are giving a clear notion how selected portfolio models perform in real-world environment. Our study comprises CreditMetrics, CreditRisk+ and KMV model. In the first part of the thesis, our intention is to clarify theoretically main features, modeling principles and moreover we also suggest hypotheses about strengths/drawbacks of every scrutinized model. Subsequently, in the applied part we test the models in a lab-environment but with real-world market data. Noticeable stress is also put on model calibration. This enables us to con firm/reject the assumptions we made in the theoretical part. In the very end there follows a straightforward general overview of all outputs and a conclusion.

This thesis deals with credit risk and selected methods of its evalua- tion. It is focused on assumptions, calculation methods, results and specifics of the CreditMetrics and the CreditRisk+ models. The CreditRisk+ model analytically determines the portfolio credit losses distribution that is caused by defaults of counterparties. In the CreditMetrics model, the credit migration risk is addition- ally considered and the future portfolio value distribution is calculated using the Monte Carlo simulation. The third approach covered in this thesis is the Solvency II, the set of requirements proposed by the European Union for determination of regulatory capital for insurance companies. In the practical part the three ap- proaches are applied on a set of three portfolios of different credit quality. Their results, particularly the determined level of capital required to cover the risk of unexpected credit losses, are analyzed and compared.

Made available in DSpace on 2016-03-15T19:25:40Z (GMT). No. of bitstreams: 1 Rafael Felicio Mileo Neto.pdf: 1273950 bytes, checksum: 918a06ee6a1bd5351e3b400a3731cfdd (MD5) Previous issue date: 2011-03-14 / Fundo Mackenzie de Pesquisa / The susceptibility of credit market to losses encouraged the institution of regulations, such as Basel I and II Capital Accords, which stimulated the development of credit portfolio management models. The objective of this dissertation is to observe the behavior of two advanced models of credit portfolio risk, KMV, based on Merton (1974) studies, and CreditRisk+, developed by Credit Suisse Financial Products. The study aims to evaluate the performance of each model in sample credit portfolios, according to market, account and debt data of companies. Through variations in each model parameters, the model`s performances in different scenarios will also be analyzed. The research focuses specifically on loss distributions generated by the models, given the changes in the parameters during the simulations. To achieve these goals, the historical evolution of credit risk is discussed, starting with the first registered loans, in Antiquity, up until the last decade, when many international regulations to credit risk were created. / A suscetibilidade do mercado de crédito a perdas incentivou a criação de regulamentações, como os Acordos de Capital da Basileia I e II, que estimularam o desenvolvimento de modelos de gestão de portfólio de crédito. O objetivo desta dissertação é observar o comportamento de dois modelos avançados de mensuração do risco: o KMV, baseado nos estudos de Merton (1974), e o CreditRisk+, criado pela Credit Suisse Financial Products. O estudo pretende verificar seus desempenhos em amostras de carteiras baseadas em informações contábeis, de mercado e de títulos de dívidas de empresas, através de variações aplicadas nos parâmetros de cada modelo, serão realizadas, também, análises de desempenho dos modelos em diferentes cenários. A pesquisa foca, especificamente, as distribuições de perdas geradas pelos modelos. Para a conclusão desses objetivos, o risco de crédito é abordado conforme sua evolução histórica, iniciando com os primeiros registros de concessão de financiamentos, na Antiguidade, até o passado recente, quando surgiram as principais regulamentações transnacionais para o risco de crédito.

A atividade bancária envolve em suas operações diversas formas de riscos. Dentre esses riscos está o risco de crédito representado como sendo uma medida de incerteza relacionada ao recebimento de um valor compromissado concedido pela instituição financeira ao tomador de empréstimo. Nesse trabalho são apresentadas as principais metodologias de quantificação do risco de crédito como Credit Metrics, KMV, Credit Portfolio View e CreditRisk+. Esta última metodologia é aplicada a quatro portfólios de financiamentos à pessoa jurídica, evidenciando o Capital Econômico Alocado - CEA, a distribuição do risco de crédito em diferentes ramos e setores de atividade da economia e o spread necessário para cobrir as perdas esperadas e inesperadas. Após essa quantificação do risco de crédito, verifica-se, utilizando o conceito de Risk Adjusted Returno on Capital - RAROC, qual dos quatro portfólios de empréstimo bancário foi o mais rentável para a instituição financeira. / Banking operations involve several kinds of risk. Among those risks, there is one called the credit risk associated with a measure of uncertainty related to receiving pré-committed values from the financial institutions credit-takers. In this research, the main methodologies used for the quantification of credit risk are discussed: Credit Metrics, KMV, Credit Portfolio View e CreditRisk+. The later is then applied to four company-targeted lending portfolios, thus showing Allocated Economic Capital AEC, the distribution of credit risk in different sectors and industries in the economy, and the necessary spread for covering expected and unexpected losses. After this effort to quantify credit risk, proceed to check, using the concept of Risk Adjusted Return on Capital RAROC, which of the four lending portfolios proved to be more profitable for the financial institution.

The various versions of CreditRisk+ have widely been used in the financial industry. We compute the loss distribution under CreditRisk+ model by fast fourier transform technique in order to have faster and more stable results. Moreover, we link the parameters of the model to the exogenously observed variables which could be obtained from the financial markets by the use of Poisson INAR process. It is shown that the estimation of the parameters become available under this setup. This enables us to build a system that allows users to monitor and predict the banks loss characteristics without having specific and current information on banks.

本文修改信用風險商業化模型CreditRisk+, 以蒙地卡羅模擬的方式探 討若修改其中的兩個假設, 將對組合損失分配帶來何種影響。一、本文認 為不同產業間的違約不是獨立, 而應具有相關性。我們以4大產業的季違 約廠商數為應變數, 景氣因子作為自變數估計各產業違約情形與總體經 濟間的關係。二、個別公司違約損失率是一與違約機率相關的隨機變數, 而不再是常數。本文提出利用財報試算各公司、產業LGD 的方法, 並假 設產業違約損失率為Beta 隨機變數, 而其中的參數會受總體因子影響。 如此一來, 產業間的違約機率與違約損失率因總體因子的關係不再獨立, 於是個別公司的違約機率、違約損失率將具有相關性。 最後, 我們以台灣上市櫃公司中同時具有TCRI 評等資訊以及財報 的561家公司作為虛擬的放款組合, 模擬在不同總體條件下的信用損失分 配。結果顯示, 在考慮了產業間違約相關性後的損失將大於產業獨立時的 損失, 而進一步納入違約機率和損失率的相關性後, 放款組合的預期損失 與風險值也隨之提高。

A atividade bancária envolve em suas operações diversas formas de riscos. Dentre esses riscos está o risco de crédito representado como sendo uma medida de incerteza relacionada ao recebimento de um valor compromissado concedido pela instituição financeira ao tomador de empréstimo. Nesse trabalho são apresentadas as principais metodologias de quantificação do risco de crédito como Credit Metrics, KMV, Credit Portfolio View e CreditRisk+. Esta última metodologia é aplicada a quatro portfólios de financiamentos à pessoa jurídica, evidenciando o Capital Econômico Alocado - CEA, a distribuição do risco de crédito em diferentes ramos e setores de atividade da economia e o spread necessário para cobrir as perdas esperadas e inesperadas. Após essa quantificação do risco de crédito, verifica-se, utilizando o conceito de Risk Adjusted Returno on Capital - RAROC, qual dos quatro portfólios de empréstimo bancário foi o mais rentável para a instituição financeira. / Banking operations involve several kinds of risk. Among those risks, there is one called the credit risk associated with a measure of uncertainty related to receiving pré-committed values from the financial institutions credit-takers. In this research, the main methodologies used for the quantification of credit risk are discussed: Credit Metrics, KMV, Credit Portfolio View e CreditRisk+. The later is then applied to four company-targeted lending portfolios, thus showing Allocated Economic Capital AEC, the distribution of credit risk in different sectors and industries in the economy, and the necessary spread for covering expected and unexpected losses. After this effort to quantify credit risk, proceed to check, using the concept of Risk Adjusted Return on Capital RAROC, which of the four lending portfolios proved to be more profitable for the financial institution.