A aplicação de Redes Bayesianas como modelo causal em Risco Operacional e extremamente atrativa do ponto de vista do gerenciamento dos riscos e do calculo do capital regulatorio do primeiro pilar do Novo Acordo da Basileia. Com as Redes e possível obter uma estimativa do VAR operacional utilizando-se não somente os dados históricos de perdas, mas também variáveis explicativas e conhecimento especialista através da possibilidade de inclusão de informações subjetivas. / The application of Bayesian Networks as causal model in Operational Risk is very attractive from the point of view of risk management and the calculation of regulatory capital under the first pillar of the New Basel Accord. It is possible to obtain with the networks an estimate of operational VAR based not only on the historical loss data but also in explanatory variables and expert knowledge through the possibility of inclusion of subjective information.

Submitted by Luciano Rachman (lucianora@uol.com.br) on 2013-09-03T14:15:04Z No. of bitstreams: 1 Dissertacao_Luciano_Rachman.pdf: 1167975 bytes, checksum: da1c59096eda72630b44358c1d1e0b0f (MD5) / Approved for entry into archive by Suzinei Teles Garcia Garcia (suzinei.garcia@fgv.br) on 2013-09-03T14:52:47Z (GMT) No. of bitstreams: 1 Dissertacao_Luciano_Rachman.pdf: 1167975 bytes, checksum: da1c59096eda72630b44358c1d1e0b0f (MD5) / Made available in DSpace on 2013-09-03T15:05:37Z (GMT). No. of bitstreams: 1 Dissertacao_Luciano_Rachman.pdf: 1167975 bytes, checksum: da1c59096eda72630b44358c1d1e0b0f (MD5) Previous issue date: 2013-08-07 / As perdas trabalhistas nas Instituições Financeiras representam um valor considerável que devem ser consideradas no modelo de capital regulatório para risco operacional, segundo Basileia. A presente dissertação demonstra uma forma de mensurar o risco às quais as Instituições Financeiras estão expostas nesse tipo de perdas. Diversos tipos de distribuições são analisados conforme sua aderência tanto na frequência como na severidade das perdas. Para os valores de frequência, foi obtida uma amostra de dados real, enquanto para a severidade foram utilizados valores obtidos de relatórios de instituto de pesquisa que serviram de insumo para os cálculos de ações trabalhistas conforme legislação brasileira vigente na CLT (Consolidação das Leis do Trabalho). / According to Basel, the labor losses in Financial Institutions represent a substantial value that should be regarded in the model of regulatory capital for operational risk. This dissertation demonstrates a way to measure the risk to which Financial Institutions are exposed to in this type of loss. Several types of distributions are analyzed according to their adherence both in frequency and severity of losses. For frequency values, it was obtained a sample of actual data, whilst for the severity were used values given from reports of research institute which served as an input for the calculations of labor actions according to the present Brazilian legislation in CLT (Consolidation of Labor Laws).

[pt] O gerenciamento de risco em instituições bancárias, mais do que mera imposição das agências reguladoras distingue-se como fator de sucesso na melhoria dos processos, aumentando o resultado financeiro. Após o Acordo da Basiléia, a gerência de riscos de mercado e de crédito, cuja atuação se dá sobre as receitas, passou a ser realizada. Entretanto, alguns riscos atuam sobre as despesas, destacando-se o operacional, que é o risco de perdas oriundas de problemas com controles internos, sistemas, pessoas e eventos externos. O objetivo deste trabalho foi elaborar uma revisão abrangente da literatura e um protótipo de sistema computacional que permite medir o VaR do risco operacional de uma unidade de risco, utilizando o Modelo de Distribuição de Perdas (LDA), e aplicar modelos causais que expliquem estas perdas. Este protótipo é uma aplicação Internet/intranet desenvolvida na linguagem ASP e utilizou o MS-Access como banco de dados. Para os cálculos estatísticos, implementou-se uma interface de comunicação aplicação/MATLAB. A revisão da literatura objetivou a familiarização com conceitos básicos de risco operacional descritos pelo Comitê da Basiléia. Adicionalmente, apresentou detalhes técnicos para implementação do LDA, tais como Distribuição de Freqüência e de Severidade, métodos para determinação da distribuição de perdas operacionais e construção da base de dados de perdas. Independente das particularidades institucionais, esse protótipo permite a visualização das providências estratégicas e operacionais a serem tomadas para implementação e implantação de um sistema similar. Marca um ponto de partida para o desenvolvimento de um produto abrangente de gerenciamento de risco operacional nas mais variadas instituições e segmentos de mercado. / [en] The risk management in financial institutions, more than just an imposition of the regulatory agencies, represents a success factor in the processes enhancement, elevating the financial results. After Basel Accord, credit and market risks management, which acts over earnings, were implemented. However, some risks are associated to the expenses, such as the operational risk, related to the losses from internal control, systems, human and external events problems. The aim of the present study was the elaboration of an extensive literature review and the development of a computation system prototype able to measure the operational risk VaR of a risk unit, using the Loss Distribution Approach (LDA) and to apply causal models that explain these losses. This prototype is an Internet/intranet application developed in ASP language, using MS-Access as database. For statistical evaluation, an interface between the application and MATLAB was implemented. The literature review pretended to give a better understanding of the basic concepts of operational risk described by the Basel Committee. In addition, it presented technical details for LDA implementation, such as Frequency and Severity Distribution, methods for the distribution of the operational losses determination and losses database construction. Independent of institutional peculiarities, this prototype allows the observation of strategic and operational providences to be taken for implementation and implantation of a similar system. It determines a startingpoint in the development of an operational risk management product valuable in several institutions and market segments.

放款的利息收入雖是商業銀行主要之獲利來源, 但借貸行為卻同時使得銀行承受著違約風險。銀行應透過風險管理方法, 計算經濟資本以提列足夠準備來防範預期以及未預期損失。 另外, 若銀行忽略違約行為之間的相關性, 將有可能低估損失的嚴重性。因此, 為了在考量違約相關性下提列經濟資本, 本文由 Merton (1974) 模型出發, 以信用風險因子模型判定放款對象是否違約, 進而決定銀行面對的整體損失為何。 為簡化分析, 本文假設違約損失率 (loss given default) 為 100%。 再者, 為加強相關性, 本文亦將違約傳染性加入因子模型並比較有無傳染性效果時, 模型所計算出的損失孰輕孰重。 而在決定違約與否時, 須利用來自移轉矩陣上的無條件違約機率, 然信評機構所發布之移轉矩陣概遺漏諸多訊息, 依此, 本文以多期連續的移轉矩陣修正之並得到另一不同的無條件違約機率。 最後, 以臺灣的 537 家上市櫃公司作為資產組合, 經由蒙地卡羅模擬得到兩個因子模型的損失分配, 我們發現具有傳染性效果存在時, 預期損失和非預期損失較大且損失分配也較為右偏。 / Despite interest income from loans is a major profit contributor for commercial banks, lending inevitably makes banks bear default risks. For the sake of avoiding expected and unexpected losses, risk management methods ough to be employed by banks to meet the ecomical capital requirement. Besides, loan loss severity may very well be underestimated if the correlation between default events is disregarded. Therefore, in order to calculate economical capital when taking default correlation into account, we start from Merton (1974) model, and identify if loans will be in default via facor models for portfolio credit risk and portfolio losses can then be detemined. To simplify our analysis in this paper, loss given default is assumed to be 100%. To intensify correlation, default contagion is, moreover, introduced to our factor model and we investigate which model results in larger losses as well. When determining default, we have to utilize rating transition matrices to obtain unconditional probability of default. Transition matrices published by credit rating agencies, however, have embedded drawback of insufficient information. We correct this flaw by means of another transition matrix based on continuous-time observations and produce different unconditional probability of default. Through Monte Carlo simulation, loss distributions are calibrated respectively from the two factor models under portfolio of 537 Taiwan listed and OTC companies. We find that expected and unexpected losses are larger and loss distribution is more right-skewed when infectious effects exsit.

The debate on quantitative operational risk modeling has only started at the beginning of the last decade and the best-practices are still far from being established. Estimation of capital requirements for operational risk under Advanced Measurement Approaches of Basel II is critically dependent on the choice of risk measure, which quantifies the risk exposure based on the underlying simulated distribution of losses. Despite its well-known caveats Value-at-Risk remains a predominant risk measure used in the context of operational risk management. We describe several serious drawbacks of Value-at-Risk and explain why it can possibly lead to misleading conclusions. As a remedy we suggest the use of coherent risk measures - and namely the statistic known as Expected Shortfall - as a suitable alternative or complement for quantification of operational risk exposure. We demonstrate that application of Expected Shortfall in operational loss modeling is feasible and produces reasonable and consistent results. We also consider a variety of statistical techniques for modeling of underlying loss distribution and evaluate extreme value theory framework as the most suitable for this purpose. Using stress tests we further compare the robustness and consistency of selected models and their implied risk capital estimates...

本文主要研究動態回復率之建構。並搭配使用機率勺斗法，將資產之離散損失分配建構出合成型擔保債權憑證分劵損失分配。歸納出離散動態回復率對合成型擔保憑證分劵之風險承擔與信用價差變化。本文發現在動態回復率中，即使在相同條件下有一樣預期損失，能使其債權群組損失分配之標準差較固定回復率小，且可使投資組合巨額損失部份產生厚尾分配現象。動態回復率對各分劵面臨共同存活與違約機率具有緩和或增強分劵承擔風險之作用。在單因子高斯連繫結構靜態違約下，透過隨機回復率能增加動態系統性風險因子之描繪。類似於將系統風險因子分配由標準常態分配改成t分配或是債權群組間違約相關係提高。

Our interest in this thesis is first to combine the different measurement techniques for operational risk in financial companies, and we highlight more and more the consequences of estimation risk which is treated as a particular part of operational risk. In the first part, we will present a full overview of operational risk, from the regulatory laws and regulations to the associated mathematical and actuarial concepts as well as a numerical application regarding the Advanced Measurement Approach, like Loss Distribution to calculate the capital requirement, then applying the Extreme Value Theory. We conclude this first part by setting a scaling technique based on (OLS) enabling us to normalize our external data to a local Lebanese Bank. On the second part, we feature estimation risk by first measuring the error induced on the SCR by the estimation error of the parameters, to having an alternative yield curve estimation and finishing by calling attention to the reflections on assumptions of the calculation instead of focusing on the so called hypothesis "consistent with market values", would be more appropriate and effective than to complicate models and generate additional errors and instability. Chapters in this part illustrate the estimation risk in its different aspects which is a part of operational risk, highlighting as so the attention that should be given in treating our models

從民國87年所通過的強制汽車責任保險，可發現我國強制汽車責任險的理賠上限與承保範圍等有了重大的改變，造成汽車任意責任險的計算費率必須要重新估算，然而國內對此方面的文獻探討卻著墨不多，因此學生將會針對任意汽車體傷責任險費率釐算詳細加以探討。 而若要重新估計任意汽車責任險首先要做的工作就是要收集完整正確的損失資料，不過由於損失資料的收集相當困難，因此只能透過模擬的損失資料進行任意責任險的費率釐算。而在有模擬的損失資料情況下我們就可透過損失分佈理論進行下列的分析： (1)透過損失資料的特性推估任意汽車責任險可能之損失分佈為Lognormal 分佈。 (2)透過最大概似估計法與特殊法推估Lognormal分佈之參數，並且採用負對數蓋似函數選擇最佳之估計參數。 (3)透過與強制汽車責任險預期損失與汽車任意體傷責任險預期損失之比例關係，釐算汽車任意體傷責任險之純保費。 (4)建立兩種損失趨勢函數，並透過此兩趨勢函數計算汽車任意體傷責任險之高保額係數。 (5)透過上述步驟之計算結果與現行實施之汽車任意體傷責任險費率作比較，以探討是否現行費率是否有超收或不足的現象。 總之，希望此篇論文能夠對未來的汽車任意責任險之費率釐算與保險司費率監督有所幫助。 / Cause the Legislation Yuan passed the compulsory auto liability insurance bill in 1998, we must have a new actuarial pricing of voluntary auto third party insurance. However, all domestic insurers haven’t revised the rate because the absence of the empirical loss data. In addition, only a fewer researches have focused on the actuarial model of this type of insurance. In this paper, we will investigate the pure premium calculation of the voluntary auto insurance, and outline the appropriate model construction procedures. The data we use are not empirical loss data, we calculate the pure premium by the simulated data. The procedures of this study are summarized in the following: (1) Find the possible loss distribution of voluntary auto third party insurance policy. (2) Estimate the parameters of the loss distribution by the maximum likelihood estimate method and the special method of lognormal distribution. (3) Calculation the pure premium of voluntary auto third party insurance. (4) Calculation the increased limits factor(ILF)by two trend functions, and compare the results of two trend functions. (5) Finally, we examine the gross premiums of the voluntary auto third party insurance and compare our results with the actual voluntary auto liability insurance premiums. Altogether, we hope that this paper could be beneficial to the actuaries and also provide suggestions for the government surveillance.

Only in English.

The aim of this master thesis is to design and analyse a FSCW PMaSynRM (Permanent Magnet assisted Synchronous Reluctance Motor) for industrial applications. The design process includes analytical calculations (initial design and PM amount minimization) and nite element method (FEM) based design optimization. An overcompensated design is proved to be advantageous for a 10-pole reluctance motor. A comparative analysis with other rotor topologies was made, where motor performance, temperature e ects and production costs are taken into account. Detailed curves which describe eciency, power factor and current with respect to ambient temperature are studied for the proposed motor designs at di erent working points. The demagnetization risk is also taken into account and the safe working temperature ranges have been dened for all the considered motors. The results show that the initial motor design with 10 poles/12 slots PMaSynRM with NdFeB magnets has poor performance in terms of eciency and power factor, with huge amount of PM inserted. This is mainly due to the lack of reluctance torque for this relatively higher number of poles solution. Moreover, it has been found in literature and conrmed in this investigation that this negative e ect for the 10-pole motor is amplied due to the presence of the concentrated winding. Indeed, it is shown by simulations that the motor performance is improved by employing 8 poles/12 slots PMaSynRM conguration with a relatively lower NdFeB magnet amount, thanks to the improved rotor anisotropy. The 10 poles/12 slots interior permanent magnet (IPM) and surface mounted permanent magnet (SMPM) topologies present higher performance due to the e ective utilization of PM, mainly or completely producing the torque. Hence, IPM and SMPM do not su er the lack of anisotropy. / Syftet med detta examensarbete ar att utforma och analysera en FSCW PMaSynRM (Permanent Magnet assisted Synchronous Reluctance Motor) for industriella applikationer. Designprocessen omfattar analytiska berakningar (ursprungliga konstruktion och PM belopp minimering) och nita elementmetoden (FEM) baserad design optimering. En overkompenserad design visat sig vara fordelaktigt for en 10-polig reluktansmotor. En jamforande analys med andra rotor topologier gjordes, dar motor prestanda, temperature ekter och produktionskostnader beaktas. Detaljerade kurvor som beskriver e ektivitet, e ektfaktor och strom med avseende pa omgivningstemperatur studeras for de foreslagna motorn for vid olika arbetspunkter. Den avmagnetisering risken ocksa beaktas och sakerhetstemperaturomraden har denierats for alla ansag motorerna. Resultaten visar att den initiala motordesign med 10-polig/12 spar PMaSynRM med NdFeB magneter har daliga e ektivitet och e ektfaktor, med enorma mangder PM insatt. Detta ar framst pa grund av bristen pa reluktansvridmomentet for denna relativt hogre poltal losning. Dessutom har man funnit i litteraturen och bekraftat i denna unders okning att denna negativa e ekt for 10-polig motorn forstarks pa grund av narvaron av den koncentrerade lindningen. Faktum ar att det framgar av simuleringar att motorprestanda forbattras med en 8-polig/12 spar PMaSynRM konguration med en relativt lagre NdFeB magnet belopp, tack vare den forbattrade rotor anisotropi. Den 10-polig/12 spar interior permanentmagnet (IPM) och ytmonterade permanent magnet (SMPM) topologier presentera hogre prestanda tack vare ett e ektivt utnyttjande av PM och deras produktion vridmoment, huvudsakligen eller helt anfortrotts PM effekten.