|
|
|
Abstracts Wednesday, January 31, 2001, at 13:30: Markus Fulmek, Stefan Pichler, Johanna Gaier, Christopher Summer This introductory course has been designed to cover and clearly explain the theory and practice of credit risk management. After explaining the general concept of credit risk we will first look at legal aspects and today's banking regulation. Then we will treat the issues of credit valuation, portfolio theory and the measurement of risk by explaining and discussing several models. These include rating based models (e.g. CreditMetrics), option-type models (e.g. KMV Model) and insurance models (e.g. Credit Risk Plus). Special emphasis will be put on a comparison of the new internal models showing their advantages and disadvantages (implementation, data calibration,...). Only this first day's program will be given in German.
Thursday, February 1, 2001, at 8:30: Andreas Ittner The keynote address will provide an overview of the new capital adequacy framework and demonstrate the key role of credit risk and its management within this new framework. The talk will focus on a discussion of the main implications of the proposed new regulations on banks and on banking supervision. This discussion will highlight the importance of a comprehensive and integrated concept of credit risk management. The talk will conclude with an outlook for the likely future agenda for banks and supervisors, as well as for research in order that the new capital accord can be implemented as effectively as possible and further development of credit risk management methods and tools undertaken.
Thursday, February 1, 2001, at 9:00 Walter Mussil
Thursday, February 1, 2001, at 10:30: Stefan Hohl Presentation of the Basel Committee's new approach based on internal ratings to regulatory capital that more accurately reflects a bank?s individual risk profile. Discussion of the key elements of the more risk sensitive IRB approach for different portfolios in the context of the foundation and advanced IRB aproach and its associated regulatory minimum requirements.
Thursday, February 1, 2001, at 11:30: Franz Partsch The presentation will begin with an overview of the provisions on credit risk in the new capital adequacy framework. Although data on credit risk is relatively scarce at present the main focus of the talk will be on identifying the more important issues for Austrian banks arising from the combination of banking statistics and large exposures risk data with credit quality information from rating agencies and other sources. As a first conclusion from this exercise we will endeavour to sketch out a (highly tentative) "road map" of the new capital accord for Austrian banks and banking supervisors.
Thursday, February 1, 2001, at 14:00: Gerhard Stahl The application of internal ratings plays a central role within the current proposal of the revised Basel Accord. Therefore the need for methods to analyse the reliability of these basic inputs is obvious. The talk will consists of three parts. The first and introductory one gives a critical review of theoretical papers that tackle the backtesting problem in the framework for credit models. The other two are devoted to a first empirical analysis of two extensive databases from two major German banks. The first one is an on-going joint work with D. Lando that analyses internal ratings of corporate firms on the basis of intensity models. The second one - an also on-going joint work with R. Kiesel - analyses country ratings on the basis of well-known cohort methods. The current discussion of backtesting ratings - or credit risk models in general - emphasizes the lack of a sufficient data history as a major drawback compared to market risk models. The talk will discuss those problems that will emerge even if a large set is available.
Thursday, February 1, 2001, at 15:00: Markus Krall Introduction: Requirements on the abilities of Rating-tools in a wider context of bank risk management. Various techniques and steps employed for developing rating-algorithms meeting the required qualities. Example: The Baetge-Oliver Wyman Rating Standard Conclusion: Future Developments
Thursday, February 1, 2001, at 16:15: Oliver Blümke In many emerging markets, country risk is not neglectable while evaluating the credit risk of a single obligor or a whole loan portfolio, as can be seen by the Emerging Markets crises in the 1990's (Mexico 1994, Asia 1997, Russia 1998). In theory a BB rated emerging market obligor should have the same default probability than an obligor in for example the U.S., incorporating already the country risk. Therefore a portfolio of 100 BB rated emerging market obligors should have the same probabilities of the number of defaults than a portfolio of 100 BB rated U.S. obligors. However, past has shown that in distressed situations for the sovereign, defaults of otherwise creditworthy obligors can occur. Therefore in such situations, default correlations tend to behave completely different, than in normal times. In view of how country risk can be defined and how it is already included in many credit portfolio models an attempt will be presented how to implement such sovereign crises in a credit portfolio model.
Thursday, February 1, 2001, at 17:15: Ludger Overbeck In the talk we present the basic probabilistic concepts in modeling the credit risk in large portfolios. Then a new risk capital allocation scheme based on the notion of coherent risk measures is constructed and compared with classical approaches based on variance/covariance analysis. In the following part of the talk we show how this model can be applied to the valuation of CLO (Collaterized Loan Obligations) and similar structured transactions. In this context we also propose some concepts to model the default time as a first hitting time of a simple transformation of a multivariate correlated Brownian motion. Then we comment on two statistical questions, namely the estimation of correlation and the validation of credit risk models.
Friday, February 2, 2001, at 8:30: Darrell Duffie This talk addresses the risk analysis and market valuation of collateralized debt obligations (CDOs). We illustrate the effects of default correlation and over-collateralization for the market valuation and risk of CDOs. We emphasize several issues:
The talk is based on joint work with Nicolae Garleanu.
Friday, February 2, 2001, at 10:15: Mark H.A. Davis 'Reduced form' credit risk models are those in which a default time is characterized by a 'hazard rate' process that can be calibrated to observed term structures of credit spreads. The hazard rate process in fact looks very much like a second factor in a short-rate based model for the term structure of interest rates, and there is a large literature on modelling and pricing in this framework. When considering products that depend on the credit performance of two or more entities, modelling the individual default times is far from the whole story: extra information must be provided to characterize the joint default distribution. This talk will survey various ways in which this can be done, including a characterization in terms of copula functions and 'contagion models' that specify mechanisms for joint default.
Friday, February 2, 2001, at 11:30: Josef Zechner This paper explores the effect of alternative option models on implied expected default frequencies. In the traditional Merton model used for example by KMV default occurs at the maturity of the debt if the firm value falls below the face value of debt. In a more general framework default can occur at any point in time either because debtholders may close down the firm or because equityholders find it in their own interest to default. We show that the precise nature of the default decision is an important determinant of the resulting expected default probability.
Friday, February 2, 2001, at 13:45: Peter Schaller Application of VAR-techniques to credit portfolios and the resulting movement towards an integrated view on market and credit risk has triggered a process of mutual learning between market and credit risk management. We analyse some aspects of this process with special empahsis on the following topics:
Friday, February 2, 2001, at 14:55: Michael Zerbs Capturing Interrelationships Between Market Risk and Credit Risk in the Mark-to-Future Framework
Friday, February 2, 2001, at 16:15: Christopher C. Finger Most models of portfolio credit risk rely on Monte Carlo simulations in order to calculate risk statistics. However, the typical parameters used in credit models (for instance, very small default probabilities) as well as the typical applications (large portfolios) and outputs (capital estimates at high levels of confidence) present particular problems for a direct Monte Carlo approach. In this talk, I will discuss some of the applications of credit models and the challenges they pose to Monte Carlo techniques. I will then present a number of approaches to dealing with these challenges. Among the topics I will cover will be simple techniques for handling large portfolios, hybrid approaches combining Monte Carlo with analytic solutions, and applications of importance sampling to mitigate simulation noise.
Friday, February 2, 2001, at 17:45: Rainer Fuhrmann The transformation from a traditional to a state-of-the-art credit risk management function is by no means just a question of employing the right formula and technology. The presentation will describe and addresses the soft but complex issues which arise when implementing new credit rating and credit limit systems, and credit risk portfolio models to establish a modern credit risk function.
|
back to wednesday's program
next abstract
up to the menu