Zugl.: Vallendar, Wiss. Hochsch. für Unternehmensführung, Diss., 2005

Thesis (doctoral)--Universität St. Gallen, 2009.

Teilw. zugl.: London, Univ., Diss., 2008

In this thesis we will deal with the term of operational risk, as it is presented in the directives Basel 2 that are mandatory for financial institutions in the European Union. The main problem is operational risk modeling, therefore, how to measure and manage it. In the first part we will look at the possibility of calculating the capital requirements for operational risk under Basel 2, mainly the calculation with the internal model. We will describe the specific procedures for the development of the internal model and we will focus on Loss Distribution Approach. The internal model will be based on modeling of loss in each risk cell separately. In the second part we will show, how to include modeling of dependence structure between risk cells to the internal model with using copulas. Finally, we will show the illustrative example, where we will see, whether the modeling of dependence leads to a reduction of the total capital requirement. Powered by TCPDF (www.tcpdf.org)

The thesis presents fundamental characteristics of survival analysis in the case of competing risks and their relationships. In the case without regression, basic nonparametric estimates and a logarithmic likelihood function for parameter estimates is given. The main focus is on Cox's proportional hazards model (PH), a model with accelerated time (AFT) and a flexible regression model (FG) are also mentioned. The identifiability of the associated survival function is solved using copulas. Basics of copula theory and the measurement of dependence by correlation coefficients (Pearson, Spearman and Kendal) are described in a separate chapter. A substantial part of the theory is practically used in a generated case without regression.

This thesis is focused on the problems of the prediction of financial situation of non-financial companies in the Czech Republic and the European Union with shares publically traded on the stock exchange. It is based on the original accounting and structural models of credit risk which enable to enumerate the probability of default. The application of Merton, Altman and Ohlson model in their basic theoretical form for the determination of the probability of default or bankruptcy is dealt with in detail. Modern multi-dimensional econometric and simulation approaches enabled to extend the original Merton model, adding the theory of multi-dimensional D-Vine copulas which project the dependency between the development on the financial market into the structure of dependency of company assets of companies, which serve as a significant indicator for the calculation of the probability of default. Next steps proceed from the creation of prediction models based on the SVM method and decision trees for the purpose of classification of company financial difficulties which are represented by the debt index or liquidity index. Different forms and numbers of input variables are used for this purpose: accounting, market and combined data together with the derived probabilities of default, or more precisely of financial distress.

This cumulative dissertation is organized as follows: After the introduction, the second chapter, based on “Asymptotic independence of bivariate order statistics” (2017) by Falk and Wisheckel, is an investigation of the asymptotic dependence behavior of the components of bivariate order statistics. We find that the two components of the order statistics become asymptotically independent for certain combinations of (sequences of) indices that are selected, and it turns out that no further assumptions on the dependence of the two components in the underlying sample are necessary. To establish this, an explicit representation of the conditional distribution of bivariate order statistics is derived. Chapter 3 is from “Conditional tail independence in archimedean copula models” (2019) by Falk, Padoan and Wisheckel and deals with the conditional distribution of an Archimedean copula, conditioned on one of its components. We show that its tails are independent under minor conditions on the generator function, even if the unconditional tails were dependent. The theoretical findings are underlined by a simulation study and can be generalized to Archimax copulas. “Generalized pareto copulas: A key to multivariate extremes” (2019) by Falk, Padoan and Wisheckel lead to Chapter 4 where we introduce a nonparametric approach to estimate the probability that a random vector exceeds a fixed threshold if it follows a Generalized Pareto copula. To this end, some theory underlying the concept of Generalized Pareto distributions is presented first, the estimation procedure is tested using a simulation and finally applied to a dataset of air pollution parameters in Milan, Italy, from 2002 until 2017. The fifth chapter collects some additional results on derivatives of D-norms, in particular a condition for the existence of directional derivatives, and multivariate spacings, specifically an explicit formula for the second-to-last bivariate spacing. / Diese kumulative Dissertation ist wie folgt aufgebaut: Nach der Einleitung wird im zweiten Kapitel, welches auf “Asymptotic independence of bivariate order statistics” (2017) von Falk und Wisheckel beruht, die asymptotische Abhängigkeitsstruktur von bivariaten Ordnungsstatistiken untersucht. Dazu wird eine explizite Darstellung der bedingten Verteilung einer bivariaten Ordnungsstatistik hergeleitet. Kapitel 3, basierend auf “Conditional tail independence in archimedean copula models” (2019) von Falk, Padoan und Wisheckel, zeigt, dass unter schwachen Anforderungen an den Generator einer Archimedischen Copula die übrigen Komponenten unabhängig werden, wenn man auf eine davon bedingt. Das insbesondere auch dann, wenn die Komponenten ohne die Bedingung abhängig waren. Die theoretischen Erkenntnisse werden anhand von Simulationsergebnissen verdeutlicht. “Generalized pareto copulas: A key to multivariate extremes” (2019) von Falk, Padoan und Wisheckel liefert Kapitel 4. Es wird ein nichtparametrischer Ansatz vorgestellt um die Überschreitungswahrscheinlichkeit eines Zufallsvektors über einen festen, hohen Schwellenwert zu schätzen, wenn dieser einer verallgemeinerten Pareto Copula folgt. Das Verfahren wird in den theoretischen Rahmen eingebettet, anhand einer Simulation validiert und auf Luftverschmutzungsdaten in Mailand, Italien, von 2002 bis 2017 angewendet. Im fünften Kapitel werden einige weitere Ergebnisse gesammelt: es geht um Ableitungen von D-Normen, insbesondere um eine Bedingung, die die Existenz der Richtungsableitungen sicherstellt. Außerdem werden multivariate Spacings thematisiert.

The determination of the sufficient amount of economic capital and its allocation to the business lines is the key issue for insurance companies. In this thesis we introduce two methods of aggregating economic capital. One is based on linear correlation and the second deals with copulas. A multitude of allocation principles have been proposed in the literature. We choose those which are the most used in practice and compare advantages and disadvantages of their application. The last chapter is devoted to the numerical examples of capital aggregation and allocation principles. 1

Claims reserving and claims process estimation present classical problems in general insurance. The overall reserves are often determined under the assumption of independence among the lines of business. Though, recently modelling of the dependence among multiple lines of business has become crucial issue of reserving process. In this context, copulae provide a useful tool to construct models which go beyond the classical ones in terms of dependence structure. This thesis deals, in particular, with the copula regression model, its properties and possible applications in general insurance. This approach combines GLM modelling of margins and then expressing the dependence structure using copula. The theoretical methods are illustrated on a real dataset.

In non-life insurance, the independence between the number and size of claims is usually assumed. However, this thesis shows that the assumption of independence can be omitted. We deal with the dependency modeling between frequency and severity of claims. For including the dependence to the total claims model, we consider two methods. The first method uses generalized linear models and the second method used in the thesis is based on dependence modeling by copulas. We also perform a model with independent frequency and severity of claims. This model is compared with the described methods in the simulation part of the thesis. We include dependency on explanatory (rating) variables in all of these models. 1