This file was created by the TYPO3 extension publications --- Timezone: CEST Creation date: 2026-05-27 Creation time: 22:59:01 --- Number of references 127 We propose a novel framework for modeling large dynamic covariance matrices via heterogeneous autoregressive volatility and correlation components. Our model provides direct forecasts of monthly covariance matrices and is flexible, parsimonious and simple to estimate using standard least squares methods. We address the problem of parameter estimation risks by employing nonlinear shrinkage methods, making our framework applicable in high dimensions. We perform a comprehensive empirical out-of-sample analysis and find significant statistical and economic improvements over common benchmark models. For minimum variance portfolios with over a thousand stocks, the annualized portfolio standard deviation improves to 8.92% compared to 9.75–10.43% for DCC-type models. article 2025 7 02 10.1016/j.jbankfin.2025.107505 Journal of Banking & Finance 178 Elsevier 107505 https://epub.uni-regensburg.de/id/eprint/77199 Igor Honig Felix Kircher This study is the first to examine the real estate-specific determinants of REIT bond risk premia. Using a dataset of 33,857 U.S. REIT bond yield spreads and 24 explanatory variables, we predict REIT bond yield spreads with a non-parametric artificial neural network algorithm and interpret the model’s predictions using the explainable machine learning method Accumulated Local Effect Plots (ALE). We report evidence of a direct real estate factor for U.S. REIT bond yield spreads proxied by real estate market total return and REIT property type. In addition, we find a property-type diversification risk premium for REIT bonds, indicating that there is no economic benefit in the form of lower cost of bond debt for most property-type diversification at the REIT-level. We argue that this is due to higher management and valuation complexity of diversified REIT portfolios. This study’s findings have relevant implications for REIT portfolio strategy and REIT capital structure decisions, as we show that specialized REITs generally have lower bond debt costs compared to diversified REITs. Moreover, a better understanding of the drivers influencing REIT bond risk premia helps investors to effectively manage bond portfolio risks. article 2025 6 12 1573-045X,0895-5638 10.1007/s11146-025-10018-7 Journal of Real Estate Finance and Economics Springer Nature https://epub.uni-regensburg.de/id/eprint/76518 Jakob Kozak Cathrine Nagl Maximilian Nagl Eli Beracha Wolfgang Schäfers Real Estate Investment Trust (REIT) returns and volatility have been extensively studied, yet typically in isolation from each other. Given that returns and volatility are generally connected in the eyes of investors, we simultaneously analyze the drivers of REIT returns and volatility over the modern REIT era (1991–2022) using an eXtreme Gradient Boosting (XGBoost) machine learning algorithm. We enhance transparency and utility through the application of explainable artificial intelligence (XAI) techniques, particularly SHapley Additive exPlanations (SHAP) and Accumulated Local Effects (ALE), which unpack the decision-making process of the model. Our analysis reveals that while no single feature consistently dominates, the influence of various drivers fluctuates significantly over time. Notably, the importance of macroeconomic indicators generally diminishes, while REIT-specific characteristics become more influential during the sample period. Furthermore, market cycles (macroeconomic shocks) cause large deviations from otherwise long-run patterns. However, during these times of economic uncertainty, drivers of risk and return correlate more strongly in comparison to times of economic stability. Lastly, we find non-linearities in the way the drivers influence returns and volatility. These insights have significant implications for investors, policymakers, and researchers as they navigate the evolving landscape of real estate investments. article 2025 4 01 1573-045X,0895-5638 10.1007/s11146-025-10016-9 The Journal of Real Estate Finance and Economics Springer https://epub.uni-regensburg.de/id/eprint/76513 Hendrik Jenett Cathrine Nagl Maximilian Nagl S. McKay Price Wolfgang Schäfers Purpose – As blockchain-based virtual worlds gain prominence within the emerging metaverse and Web3, numerous global companies and investors are buying purely virtual land to explore new business potentials and capitalize on digital assets. Given the similaritiesto physical real estate, thisstudy examinesthe dynamics of the secondary market for virtual land and relates its returns to those of physical real estate. Design/methodology/approach – Using transaction-level data from a prominent virtual land platform, the authors construct a virtual land market index based on repeat sales index methodology from traditional real estate studies. Wavelet coherence analysisis employed to examine the dynamic correlation between virtual land and various physical real estate marketreturns. The determinants of this correlation are estimated using stepwise regression analysis. A portfolio analysis explores the implications of adding virtual land to traditional asset portfolios. Findings – The correlation between virtual and physical real estate market returns is generally low, reaching its lowest during the Covid-19 lockdowns from 2020 to 2022. It spikes during acute economic turmoil such as the initial Covid-19 outbreak or interest rate change announcements. The correlation is primarily driven by consumer and economic climate, the price ofthe virtual economy token and investor attention. Portfolio analysis indicates that virtual land can enhance risk-adjusted returns within a traditional portfolio, particularly when added to a commercial real estate portfolio. Research limitations/implications – This study examines a single virtual land market, despite it being the oldest and one of the largest. Given the rapidly evolving nature of virtual worlds, it is crucial to further test the results and include new virtual land platforms as they emerge. Practical implications – The findings provide actionable insights on portfolio implications for investors seeking alternative real-estate-like assets in the digital space. Additionally, this study offers strategic guidance for entering the metaverse, including a comprehensive overview of established virtual presences. Originality/value – With the advancing digitization of real estate markets, this study is the first to explore the correlation between market returns of virtual land in the metaverse and traditional physical real estate. The findings provide valuable empirical insights for investors, policymakers, entrepreneurs and companies interested in the intersection of digital and traditional property markets. article 2025 1 30 10.1108/JERER-04-2024-0025 Journal of European Real Estate Research Emerald https://epub.uni-regensburg.de/id/eprint/76516 Heiko Leonhard Maximilian Nagl Wolfgang Schäfers This cumulative dissertation proposes and evaluates new approaches for improving the out-of-sample performance of mean-variance optimized portfolios under parameter estimation risks. Parameter estimation risk is a key reason why applications of portfolio theory do not deliver the expected performance in practice. It arises from not knowing the true parameters that describe the behavior of future asset returns. The research in this dissertation contributes to improving the understanding of parameter estimation risks in optimal portfolio selection as well as the practical applications of portfolio theory by providing enhanced estimators of optimal portfolio weights based on empirical data. In Chapter 2 of this dissertation, a novel shrinkage estimator for optimal portfolio weights with respect to the expected out-of-sample Sharpe ratio is developed. Chapter 3 uses firm characteristics, state-of-the-art machine learning methods and covariance matrix shrinkage methods for constructing large efficient portfolios of individual stocks. Chapter 4 proposes a new approach for estimating and predicting high-dimensional time-varying covariance matrices that can be directly used for optimal portfolio selection. Taken together, the methods proposed in this dissertation expand the toolbox for academics and practitioners in the application of optimal portfolio selection models. thesis_rgbg 2025 1 23 https://epub.uni-regensburg.de/id/eprint/74539 Felix Kircher The financial sector, particularly investors, can make a significant contribution to the fight against climate change. They can drive the green transition by participating in emissions trading systems with carbon emission allowances. These allowances are suitable for responsible investing as they create economic incentives for climate action. From an investor’s perspective the question of financial performance always arises. This article investigates whether and in which way the inclusion of the liquid investable global carbon index, which includes the major and most traded European and North American cap-and-trade programs, can improve the performance of multi-asset portfolios. We model different portfolio- and investortypes, evaluate their performance, and study the contribution of carbon during different market phases and economic environments. Portfolios incorporating carbon exhibit superior performance in terms of (risk-adjusted) returns and an elevated conditional diversification benefit. The advantages for more risk-seeking investors tend to be more pronounced than those for rather conservative ones. Our findings underline the role of carbon markets as catalyst for portfolio growth and responsible investing, providing valuable insights for investors and researchers. article 2025 10.3905/jai.2025.1.250 The Journal of Alternative Investments 28 Institutional Investor 7-63 2 https://epub.uni-regensburg.de/id/eprint/78169 Alicia Billand Maximilian Nagl Daniel Rösch This cumulative doctoral thesis aims to shed light on the application of uncertainty-aware machine learning to credit risk problems. Over the last decades machine learning and artificial intelligence have become increasingly important for scientific research and practical application. Due to the superior performance compared to standard algorithms, machine learning is often the preferred choice. To ensure the reliability of these machine learning algorithms, the uncertainty associated with them is an active area of research. This is especially interesting for applications where the reliability of the model is crucial, such as in the credit risk sector. Chapter 1 focuses on one of the key parameters in credit risk, the Loss Given Default (LGD). This chapter contributes to the literature by quantifying the uncertainty in the LGD prediction by using the deep evidential regression. Chapter 2 estimates the distribution of the LGDs in a non-linear manner by introducing a novel machine learning method, the (generalized) beta regression artificial neural network (G-BRANN). Chapter 3 focuses on the important task of handling missing data. This chapter introduces a new imputation technique, GAMME, by combining predictive mean matching with accumulated local effect (ALE) plots. thesis_rgbg 2024 12 03 https://epub.uni-regensburg.de/id/eprint/59657 Matthias Nagl We synthesize the extended linear beta regression with a neural network structure to model and predict the mean and precision of market-based loss rates. We can incorporate non-linearity in mean and precision in a flexible way and resolve the problem of specifying the underlying form in advance. As a novelty, we can show that the proportion of non-linearity for the mean estimates is 14.10% and 80.37% for the precision estimates. This implies that especially the shape of the loss rate distribution entails a large amount of non-linearity and, thus, our approach consistently outperforms its linear counterpart. Furthermore, we derive trainable activation functions to allow a data-driven estimation of their shape. This is important if predictions have to be in a certain interval, e.g., (0, 1) or (0, ∞) . Conducting a scenario analysis, we observe that our estimated distributions are more refined compared to traditional models, thereby demonstrating their suitability for risk management purposes. These estimated distributions can assist financial institutions in better identifying diverse risk profiles among their creditors and across various macroeconomic states. article 2024 9 21 1436-6304,0171-6468 10.1007/s00291-024-00787-7 OR Spectrum Springer https://epub.uni-regensburg.de/id/eprint/59239 Matthias Nagl Maximilian Nagl Daniel Rösch This paper quantifies the intricacy, i.e., non-linearity and interactions of predictor variables, in explaining cryptocurrency returns. Using data from several thousand cryptocurrencies spanning 2014 to 2022, we observe a notably high level of intricacy. This provides a quantitative measure why linear models are often outperformed by machine learning algorithms in predicting cryptocurrency returns. Furthermore, we document that the intricacy in these predictions is considerably larger compared to stocks. Our analysis reveals that interactions are gaining importance over time, while individual non-linearity of the drivers is diminishing. This adds to the emerging literature on spillover effects between cryptocurrencies, traditional finance and the economy. This finding is important for investors as well as regulators as the high intricacy proposes challenges to both actors in the market. article 2024 5 07 0165-1765,1873-7374 10.1016/j.econlet.2024.111746 Economics Letters 239 Elsevier 111746 https://epub.uni-regensburg.de/id/eprint/58318 Maximilian Nagl In recent years, blockchain-based platforms such as Propy and Elysia have emerged that apply tokenization to commercial real estate. They issue real estate crypto coins, which represent a hybrid between real estate and cryptocurrencies. We investigate the return dependence of two real estate crypto coins on (1) the real estate market, as represented by the equity REIT market due to the availability of daily data, and (2) the cryptocurrency market to assess whether real estate crypto coins behave like real estate or cryptocurrencies. Using the time-varying optimal copula (TVOC) approach, we find that the dependence of real estate crypto coins on the real estate and cryptocurrency markets has changed over time. While real estate crypto coins primarily provided exposure to the cryptocurrency market in their early years, our results suggest that real estate crypto coins have become more similar to real estate as they matured. Our findings have portfolio implications for institutional investors, as they suggest that real estate crypto coins represent another asset class to be included in the real estate category. article 2023 12 06 0896-5803,2691-1175 Journal of Real Estate Research Routledge 1-29 https://epub.uni-regensburg.de/id/eprint/55219 Cathrine Nagl Maximilian Nagl Daniel Rösch Wolfgang Schäfers Julia Freybote Recent advancements in natural language processing (NLP) methods have significantly improved their performance. However, more complex NLP models are more difficult to interpret and computationally expensive. Therefore, we propose an approach to dictionary creation that carefully balances the trade-off between complexity and interpretability. This approach combines a deep neural network architecture with techniques to improve model explainability to automatically build a domain-specific dictionary. As an illustrative use case of our approach, we create an objective dictionary that can infer conflict intensity from text data. We train the neural networks on a corpus of conflict reports and match them with conflict event data. This corpus consists of over 14,000 expert-written International Crisis Group (ICG) CrisisWatch reports between 2003 and 2021. Sensitivity analysis is used to extract the weighted words from the neural network to build the dictionary. In order to evaluate our approach, we compare our results to state-of-the-art deep learning language models, text-scaling methods, as well as standard, nonspecialized, and conflict event dictionary approaches. We are able to show that our approach outperforms other approaches while retaining interpretability. article 2023 3 22 1047-1987,1476-4989 10.1017/pan.2023.7 Political Analysis CAMBRIDGE UNIV PRESS

CAMBRIDGE

1-19 https://epub.uni-regensburg.de/id/eprint/54057 Sonja Häffner Martin Hofer Maximilian Nagl Julian Walterskirchen Machine learning has increasingly found its way into the credit risk literature. When applied to forecasting credit risk parameters, the approaches have been found to outperform standard statistical models. The quantification of prediction uncertainty is typically not analyzed in the machine learning credit risk setting. However, this is vital to the interests of risk managers and regulators alike as its quantification increases the transparency and stability in risk management and reporting tasks. We fill this gap by applying the novel approach of deep evidential regression to loss given defaults (LGDs). We evaluate aleatoric and epistemic uncertainty for LGD estimation techniques and apply explainable artificial intelligence (XAI) methods to analyze the main drivers. We find that aleatoric uncertainty is considerably larger than epistemic uncertainty. Hence, the majority of uncertainty in LGD estimates appears to be irreducible as it stems from the data itself. article 2022 12 15 10.3389/fams.2022.1076083 Frontiers in Applied Mathematics and Statistics 8 Frontiers 1076083 https://epub.uni-regensburg.de/id/eprint/53278 Matthias Nagl Maximilian Nagl Daniel Rösch The calibration of financial models is laborious, time-consuming and expensive, and needs to be performed frequently by financial institutions. Recently, the application of artificial neural networks (ANNs) for model calibration has gained interest. This paper provides the first comprehensive empirical study on the application of ANNs for calibration based on observed market data. We benchmark the performance of the ANN approach against a real-life calibration framework that is in action at a large financial institution. The ANN based calibration framework shows competitive calibration results, roughly four times faster with less computational efforts. Besides speed and efficiency, the resulting model parameters are found to be more stable over time, enabling more reliable risk reports and business decisions. Furthermore, the calibration framework involves multiple validation steps to counteract regulatory concerns regarding its practical application. article 2022 7 1380-6645,1573-7144 10.1007/s11147-021-09183-7 Review of Derivatives Research 25 Springer

NEW YORK

109-136 https://epub.uni-regensburg.de/id/eprint/47893 Patrick Büchel Michael Kratochwil Maximilian Nagl Daniel Rösch For banks, credit lines play an important role exposing both liquidity and credit risk. In the advanced internal ratings-based approach, banks are obliged to use their own estimates of exposure at default using credit conversion factors. For volatile segments, additional downturn estimates are required. Using the world's largest database of defaulted credit lines from the US and Europe and macroeconomic variables, we apply a Bayesian mixed effect quantile regression and find strongly varying covariate effects over the whole conditional distribution of credit conversion factors and especially between United States and Europe. If macroeconomic variables do not provide adequate downturn estimates, the model is enhanced by random effects. Results from European credit lines suggest that high conversion factors are driven by random effects rather than observable covariates. We further show that the impact of the economic surrounding highly depends on the level of utilization one year prior default, suggesting that credit lines with high drawdown potential are most affected by economic downturns and hence bear the highest risk in crisis periods. article 2022 6 12 0964-1998,1467-985X 10.1111/rssa.12855 Journal of the Royal Statistical Society: Series A (Statistics in Society) Oxford Univ. Press

OXFORD

1-38 https://epub.uni-regensburg.de/id/eprint/52437 Jennifer Betz Maximilian Nagl Daniel Rösch Finanzinstitute spielen eine wichtige Rolle für die Stabilität des Finanzsektors. Sie bekleiden eine entscheidende Rolle als Intermediäre bei der Bereitstellung von Geld und Krediten sowie bei der Übertragung von Risiken zwischen Unternehmen. Diese Intermediärfunktion setzt die Finanzinstitute jedoch verschiedenen Arten von Risiken aus. Die Identifizierung und Messung dieser Risiken ist besonders in schwierigen Zeiten wichtig, in denen ein angeschlagener Finanzsektor zu einem Rückgang der Kreditvergabe führen kann. Vor allem in Zeiten des wirtschaftlichen Abschwungs ist die Rolle der Bereitstellung von Liquidität und Krediten wichtiger denn je. Daher ist die genaue Schätzung der Determinanten für verschiedene Risikoquellen eine äußerst wichtige Aufgabe für die Wirtschaft im Allgemeinen und für Finanzinstitute im Besonderen. In den letzten Jahrzehnten sind die Rechenleistung und die Speicherkapazitäten erheblich gestiegen, während die Kosten stark gesunken sind. Dies ermöglicht es Forschern und Praktikern, fortschrittlichere und rechenintensivere Modelle zu verwenden. Dies ist besonders wichtig für Modelle des maschinellen Lernens, aber auch für Bayesianische Modelle. Die Arbeit beleuchtet die Anwendung fortgeschrittener statistischer und maschineller Lernmethoden für das Kredit- und Marktrisikomanagement. Diese Anwendungen werden in vier unabhängigen Forschungsarbeiten behandelt. Die erste befasst sich mit fortgeschrittenen Bayesianischen Methoden, um den schwierigen Risikoparameter Exposure at Default (EAD) und sein Verhalten in Abschwungphasen zu untersuchen. Das zweite Papier konzentriert sich auf die Kombination von statistischen und maschinellen Lernmethoden, um verschiedene Aspekte der Verlustquote (Loss Given Default, LGD) zu eruieren, wobei ein besonderer Schwerpunkt auf Methoden zur Erklärbarkeit von Maschinellem Lernen liegt. Das dritte Forschungspapier wendet neuronale Netze für die Kalibrierung von Finanzmodellen an, wobei ein besonderer Schwerpunkt auf ihrem Nutzen in der Praxis liegt. Das letzte Forschungspapier befasst sich eingehend mit Nichtlinearität, die mit den Bewegungen der Aktienmärkte einhergeht. thesis_rgbg 2022 2 17 https://epub.uni-regensburg.de/id/eprint/51522 Maximilian Nagl article 2022 1059-8596,2168-8648 10.3905/jfi.2021.1.124 The Journal of Fixed Income 31 Portfolio Management Research 41-65 3 https://epub.uni-regensburg.de/id/eprint/51021 Rainer Jobst Daniel Rösch We extend the linear quantile regression with a neural network structure to enable more flexibility in every quantile of the bank loan loss given default distribution. This allows us to model interactions and non-linear impacts of any kind without the need of specifying the exact form beforehand. The precision of the quantile forecasts increases up to 30% compared to the benchmark, especially for higher quantiles which are most important in credit risk. By using a novel feature importance measure, we calculate the strength, direction, interactions and other non-linear impacts for every conditional quantile and every variable. This enables us to explain why our extension exhibits superior performance over the benchmark. Moreover, we find that the macroeconomy is up to two times more important in USA than in Europe and has large joint impacts in both regions. The macroeconomy is most important in the US, whereas in Europe collateralization is essential. (c) 2021 Elsevier B.V. All rights reserved. article 2022 1 0378-4266,1872-6372 10.1016/j.jbankfin.2021.106334 Journal of Banking & Finance 134 Elsevier

AMSTERDAM

106334 https://epub.uni-regensburg.de/id/eprint/50854 Ralf Kellner Maximilian Nagl Daniel Rösch article 2022 1755-9723,1744-6619 10.21314/JCR.2022.011 Journal of Credit Risk 18 Incisive Media 65-91 4 https://epub.uni-regensburg.de/id/eprint/52551 Rainer Jobst We consider the problem of maximizing the out-of-sample Sharpe ratio when portfolio weights have to be estimated. We apply an improved bootstrap-based estimator, and an approximative estimator derived from a Taylor series. In a simulation study and empirical analysis with 15 datasets the proposed estimators outperform the minimum variance and equally weighted portfolio strategies. Out-of-sample Sharpe ratios improve by 15 and 32 percent on average, respectively, in the empirical analysis. While effectively dealing with estimation risks, the estimators produce considerable amounts of turnover. Realized net Sharpe ratios improve by 40 percent on average when the effects of accruing transaction costs are incorporated ex-ante into estimation of portfolio weights. When adding a risk-free asset, net Sharpe ratios remain of the same magnitude and portfolio volatility does not exceed a predefined target level. (c) 2021 Elsevier B.V. All rights reserved. article 2021 8 03 0378-4266,1872-6372 10.1016/j.jbankfin.2021.106281 Journal of Banking & Finance 133 Elsevier

AMSTERDAM

106281 https://epub.uni-regensburg.de/id/eprint/50853 Felix Kircher Daniel Rösch Using access to a unique bank loss database, we find positive dependencies of default resolution times (DRTs) of defaulted bank loan contracts and final loan loss rates (losses given default, LGDs). Due to this interconnection, LGD predictions made at the time of default and during resolution are subject to censoring. Pure (standard) LGD models are not able to capture effects of censoring. Accordingly, their LGD predictions may be biased and underestimate loss rates of defaulted loans. In this paper, we develop a Bayesian hierarchical modelling framework for DRTs and LGDs. In comparison to previous approaches, we derive final DRT estimates for loans in default which enables consistent LGD predictions conditional on the time in default. Furthermore, adequate unconditional LGD predictions can be derived. The proposed method is applicable to duration processes in general where the final outcomes depend on the duration of the process and are affected by censoring. By this means, we avoid bias of parameter estimates to ensure adequate predictions. article 2021 3 12 0035-9254,1467-9876 10.1111/rssc.12474 Journal of the Royal Statistical Society, Series C 70 Wiley

HOBOKEN

619-644 3 https://epub.uni-regensburg.de/id/eprint/46338 Jennifer Betz Ralf Kellner Daniel Rösch This study analyses the level of systematic risk for US mortgage portfolio securitisations based on the variation of default rates which cannot be explained by observed deterministic factors. Systematic risk is decomposed into general systemic risk, rating-class-specific systematic risk and their covariance structure. General systematic risk sensitivities increase from lower rating classes to medium rating classes and decreases to higher rating classes. Rating-class-specific systematic risk shows an opposite pattern. The methodology provides for more accurate probability of default and Value-at-Risk forecasts. (C) 2020 Elsevier B.V. All rights reserved. article 2021 0378-4266,1872-6372 10.1016/j.jbankfin.2020.105996 Journal of Banking and Finance 122 Elsevier

AMSTERDAM

105996 https://epub.uni-regensburg.de/id/eprint/44326 Yongwoong Lee Harald Scheule Daniel Rösch This cumulative doctoral thesis amends the literature on modeling counterparty credit risk exposures. The calculation of exposure measures for counterparty credit risk is a crucial task for financial institutions as it is required for various aspects, such as derivatives valuation, assessment of capital requirements and limitation. The Global Financial Crisis (GFC) has led to a major reform of the regulatory framework for counterparty risk and the OTC derivatives market. The new regulatory requirements lead to challenges in the modeling of counterparty credit risk exposures. The thesis consists of three independent research papers, which analyze and tackle three selected issues resulting from the introduction of the new supervisory standardized approach for measuring CCR exposures (SA-CCR) and the mandatory exchange of initial margin (IM) for a multitude of non-centrally cleared derivatives. The first research paper (chapter 1) reveals weaknesses and flaws in the methodology and calibration of the SA-CCR regarding the treatment of equity options by performing a profound methodological and empirical analysis. Based on these findings, measures for improving the SA-CCR are proposed. Within the second and third research paper, innovative approaches are developed which complement the toolbox for counterparty risk management. The second research paper (chapter 2) utilizes and modifies the SA-CCR’s model framework to develop a new semi-analytical method for the generation of exposure profiles for CVA calculation. The modified SA-CCR offers a fast and simple approach for the treatment of derivatives which are not covered by advanced exposure models. The third research paper (chapter 3) introduces an innovative method for forecasting future IM requirements under ISDA-SIMM™ based on forward sensitivities. This approach utilizes information and methodological building blocks already available in a Monte Carlo framework and adopts principles from industrial just-in-time manufacturing. Hence, the approach is capable of generating fast and accurate forward sensitivities. thesis_rgbg 2020 11 03 https://epub.uni-regensburg.de/id/eprint/43882 Michael Kratochwil This paper is devoted to the parameterization of correlations in the Vasicek credit portfolio model. First, we analytically approximate standard errors for value-at-risk and expected shortfall based on the standard errors of intra-cohort correlations. Second, we introduce a novel copula-based maximum likelihood estimator for inter-cohort correlations and derive an analytical expression of the standard errors. Our new approach enhances current methods in terms of both computing time and, most importantly, direct uncertainty quantification. Both contributions can be used to quantify a margin of conservatism, which is required by regulators. Third, we illustrate powerful procedures that reduce the well-known bias of current estimators, showing their favorable properties. Further, an open-source implementation of all estimators in the novel R package AssetCorr is provided and selected estimators are applied to Moody’s Default & Recovery Database. article 2020 4 1755-2842,1465-1211 10.21314/JOR.2020.429 Journal of Risk 22 Incisive Media 1-30 https://epub.uni-regensburg.de/id/eprint/45044 Marius Pfeuffer Maximilian Nagl Matthias Fischer Daniel Rösch Considering counterparty credit risk (CCR) for derivatives using valuation adjustments (CVA) is a fundamental and challenging task for entities involved in derivative trading activities. Particularly calculating the expected exposure is time consuming and complex. This paper suggests a fast and simple semi-analytical approach for exposure calculation, which is a modified version of the new regulatory standardized approach (SA-CCR). Hence, it conforms with supervisory rules and IFRS 13. We show that our approach is applicable to multiple asset classes and derivative products, and to single transactions as well as netting sets. article 2020 1 14 1380-6645,1573-7144 10.1007/s11147-019-09165-w Review of Derivatives Research 23 Springer

NEW YORK

273-322 3 https://epub.uni-regensburg.de/id/eprint/44316 Patrick Büchel Michael Kratochwil Daniel Rösch We develop and apply a Bayesian model for the loss rates given defaults (LGDs) of European Sovereigns. Financial institutions are in need of LGD forecasts under Pillar II of the regulatory Basel Accord and the downturn in LGD forecasts under Pillar I. Both are challenging for portfolios with a small number of observations such as sovereigns. Our approach comprises parameter risk and generates LGD forecasts under both regular and downturn conditions. With sovereign-specific rating information, we found that average LGD estimates vary between 0.46 and 0.64, while downturn estimates lay between 0.50 and 0.86. (C) 2020 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved. article 2020 0169-2070,1872-8200 10.1016/j.ijforecast.2019.11.004 International Journal of Forecasting 36 Elsevier

AMSTERDAM

1073-1091 https://epub.uni-regensburg.de/id/eprint/41309 Rainer Jobst Ralf Kellner Daniel Rösch The new standardized approach for measuring counterparty credit risk exposures (SA-CCR) will replace the existing regulatory standard methods for exposure quantification. There is ongoing discussion with respect to the calibration and appropriate treatment of nonlinear products under the SA-CCR. The calibration of supervisory parameters for equity derivatives has been a particular bone of contention. Further, the SA-CCR struggles with the adequate reflection of nonstandard options. Our paper provides empirical evidence that the SA-CCR parameters are not aligned with historically observed volatilities. We explore a potential alignment of the SA-CCR with the new standardized approach for market risk (SA-TB) as well as the application of economic delta adjustments for path-dependent equity products. Our results demonstrate that an alignment of SA-CCR and the SA-TB could lead to a significantly improved risk assessment for equity derivatives. article 2020 1744-6619,1755-9723 10.21314/JCR.2020.265 The Journal of Credit Risk INCISIVE MEDIA

LONDON

https://epub.uni-regensburg.de/id/eprint/55906 Michael Kratochwil book 2020 9798617590199 Independently Published, United States https://epub.uni-regensburg.de/id/eprint/44319 Daniel Rösch Harald Scheule Resolution of non-performing loans is a key determinant of bank credit default losses. This paper analyzes macroeconomic and systematic frailty effects of the default resolution time for a sample of 17,395 defaulted bank loans in USA, Great Britain, and Canada. We find that frailties have a huge impact on the resolution times. In a representative sample portfolio, median resolution times more than double in a recession when compared to an expansion. This leads to highly skewed distributions of losses and considerable systematic risk of the bank portfolio. (C) 2017 Elsevier B.V. All rights reserved. article 2020 0378-4266,1872-6372 10.1016/j.jbankfin.2017.09.008 Journal of Banking & Finance 112 Elsevier

AMSTERDAM

1-26 https://epub.uni-regensburg.de/id/eprint/36213 Jennifer Betz Steffen Krüger Ralf Kellner Daniel Rösch Currently, the use of t statistics and p-values is under scrutiny in various scientific fields for several reasons: p-hacking, data dredging, misinterpretation or selective reporting, among others. To the best of our knowledge, this discussion has hardly reached the empirical finance community. The aim of this paper is to show how typical testing frameworks of empirical findings in finance can be fruitfully enriched by supplemental use of further statistical tools. We revisit popular studies regarding the validity of the CAPM and determine Bayesian measures for hypothesis testing, e.g., we find popular asset pricing studies might have been evaluated differently. article 2019 9 23 1544-6123,1544-6131 10.1016/j.frl.2019.101402 Finance Research Letters 38 ACADEMIC PRESS INC ELSEVIER SCIENCE

SAN DIEGO

101402 https://epub.uni-regensburg.de/id/eprint/50855 Ralf Kellner Daniel Rösch This paper adds a country specific point of view on diversification among financial markets. Previous results indicate that emerging markets may generally exhibit higher diversification effects in comparison to developed markets due to lower market linkages. However, the question remains which markets in particular may be more or less attractive among emerging and developed markets as well as in a world market setting. To the best of our knowledge, this is the first empirical study which analyzes this question beyond continental levels only. We analyze diversification by means of optimal portfolio allocations and in terms of tail risk reductions. We also include investors' tendency to overweight regional investments (home bias). Our results show that Malaysia, Switzerland, Jordan, Chile or Singapore are markets with high diversification potential. On the contrary, Argentina, Brazil and South Korea exhibit high individual risk levels in combination with high market linkages. (C) 2019 Elsevier Ltd. All rights reserved. article 2019 0261-5606,1873-0639 10.1016/j.jimonfin.2019.102064 Journal of International Money and Finance 98 Elsevier

OXFORD

102064 https://epub.uni-regensburg.de/id/eprint/48110 Ralf Kellner Daniel Rösch PurposeTorsional osteotomy of the distal femur allows anatomic treatment of patellofemoral instability and patellofemoral pain syndrome in cases of increased femoral antetorsion. The purpose of this study was to investigate the effects of distal femoral torsional osteotomy on pressure distribution of the medial and lateral patellar facet.MethodsNine fresh frozen human knee specimens were embedded in custom-made 3D-printed casts and tested with a robotic arm. Torsional osteotomy could be simulated ranging from increased femoral antetorsion of 25 degrees with a corresponding lateralization of the patella to an overcorrected value of 5 degrees of femoral antetorsion. The peak and mean lateral and medial compartment pressure was measured in 0 degrees, 15 degrees, 30 degrees, 45 degrees, 60 degrees and 90 degrees flexion beginning with neutral anatomic muscle rotation.ResultsThe medial aspect of the patella showed a significant influence of femoral torsion with an increase of mean and peak pressure in all flexion angles with progressive derotation from 15 degrees external rotation to 5 degrees internal rotation (p=0.004). The overall pressure difference was highest in near extension and stayed on a constant level with further flexion. On the lateral facet, the derotation resulted in decrease of pressure in near extension; however, it had no significant influence on the mean and peak pressure through the different torsion angles (n.s.). Unlike on the medial facet, a significant consistent increase of peak pressure from 0 degrees to 90 degrees flexion could be shown (p=0.022) on the lateral patella aspect.ConclusionDistal femoral torsional osteotomy to correct pathological femoral antetorsion leads to a redistribution of retropatellar pressure. External derotation leads to an increased peak pressure on the medial patellar facet and can impair simultaneous cartilage repair. However, as the lateral patellofemoral load decreases, it has a potential in preventing patellofemoral osteoarthritis. article 2019 0942-2056,1433-7347 10.1007/s00167-018-5165-2 Knee Surgery, Sports Traumatology, Arthroscopy 27 Springer

NEW YORK

2328-2333 7 https://epub.uni-regensburg.de/id/eprint/48507 Franz Liska Constantin von Deimling Alexander Otto Lukas Willinger Ralf Kellner Andreas B. Imhoff Rainer Burgkart Andreas Voss The accurate measurement and effective control of financial risk are of crucial importance to risk managers and regulators. However, risk measures are potentially affected by errors in the estimation of model parameters from limited samples, leading to parameter risk. The key contribution of this paper is the formulation of a general framework to hedge this parameter risk. Applying the new framework to credit portfolio modeling, we highlight the importance of parameter risk, estimation methods, and diversification effects. (C) 2019 Elsevier B.V. All rights reserved. article 2019 0378-4266,1872-6372 10.1016/j.jbankfin.2019.01.003 Journal of Banking & Finance 100 ELSEVIER SCIENCE BV

AMSTERDAM

111-121 https://epub.uni-regensburg.de/id/eprint/48914 Arndt Claussen Daniel Rösch Martin Schmelzle This paper analyses how mortgage borrower liquidity constraints and home equity drive the realized loss rates given default using loan-level data. We define defaulted loans with zero loss as cures and those with non-zero loss as non-cures. We find economically that borrower liquidity constraints and positive equity explain cure, while negative equity explains non-zero loss. The findings provide an important economic-rationale for a separation of the cure and loss processes in mortgage loss models and their applications such as loan pricing and bank capital regulation. The results have great relevance for the multi-trillion dollar mortgage industry for a more efficient capital allocation, better mortgage pricing and more forward-looking loan loss provisioning. article 2019 0895-5638,1573-045X 10.1007/s11146-019-09709-9 Journal of Real Estate Finance and Economics Springer

DORDRECHT

https://epub.uni-regensburg.de/id/eprint/41307 Hung Do Harald Scheule Daniel Rösch This cumulative doctoral thesis contributes to the broad literature on credit risk management and regulation. Following Basel II / III, financial institutions are allowed to use own estimates of central credit risk parameters to calculate their capital needs. Besides the probability of default (PD) and the exposure at default (EAD), the loss given default (LGD) is in the focus of modeling efforts in the advance Internal-Rating-Based (IRB) approach of Basel II / III. Most of the academic LGD literature applies market-based LGDs which are calculated based on market prices of defaulted debt instruments. Considering loan contracts, only workout LGDs are available in most of the cases. These are calculated based on actual recovery payments and differ considerably from market-based LGDs. First, workout LGDs are characterized by an even more extreme distributional form. Multi-modality seems to be more pronounced and bindings at no and total loss arise more commonly. Second, systematic effects among workout LGDs deviate stronger from the economic cycle. This entails a hampered identification of statistically significant / evident macroeconomic variables which is of high relevance in the light of the need for downturn estimates, i.e., estimates which reflect adverse economic surroundings. Third, workout LGDs are shaped by the resolution bias. Assuming positive dependencies of default resolution times (DRT) and LGDs, bad loan contracts are characterized by long DRTs and high LGDs. These loans are underrepresented at the end of the observation period as only LGDs of loans with short DRTs and, thus, low LGDs are observable. This might entail parameter distortions and, consequently, an underestimation of LGDs. Thus, the consideration of DRTs is crucial in the context of workout LGDs. This thesis aims to shed light on modeling DRTs and, subsequently, workout LGDs. It consists of four independent research papers. The first two papers (What drives the time to resolution of defaulted bank loans? and Macroeconomic effects and frailties in the resolution of non-performing loans) focus on DRTs, whereas, the third paper (Systematic effects among LGDs and their implications on downturn estimation) presents a sophisticated modeling approach for LGDs which aims to provide adequate downturn predictions by considering common characteristics of workout LGDs. In the fourth paper (Time matters: How default resolution times impact final loss rates), a joint modeling approach for DRTs and LGDs is developed. Hereby, effects of the resolution bias are diminished. This approach succeeds to generate adequate LGD predictions for unresolved loans – so called LGDs-in-default – and outperforms pure (standard) LGD models on an out of sample perspective. thesis_rgbg 2018 8 16 https://epub.uni-regensburg.de/id/eprint/37600 Jennifer Betz Recent credit risk literature has proposed (i) sample selection models for dependencies between the one-year Probability of Default (PD) and Loss Given Default (LGD), and (ii) multi-year approaches which are limited to default risk. This paper provides a model for the simultaneous prediction of continuous default times and multi-year LGDs. These measures are paramount to predict term structures of LGDs and Lifetime Expected Losses for the revised loan loss provisioning framework of IFRS 9 and US GAAP (current expected credit loss, CECL). The model includes a variation of copulas and corrects for sample selection bias of LGDs, which are only observed given a default event. We find empirical evidence that bonds which default closer to origination tend to generate higher LGDs. The model enables more precise estimates of Lifetime Expected Losses and prevents a severe underestimation in contrast to more restricted credit risk models. (C) 2018 Elsevier B.V. All rights reserved. article 2018 0927-5398,1879-1727 10.1016/j.jempfin.2018.04.001 Journal of Empirical Finance 47 ELSEVIER SCIENCE BV

AMSTERDAM

246-262 https://epub.uni-regensburg.de/id/eprint/37375 Steffen Krüger Toni Oehme Daniel Rösch Harald Scheule This paper develops a novel framework to model the loss given default (LGD) of residential mortgage loans which is the dominant consumer loan category for many commercial banks. LGDs in mortgage lending are subject to two selection processes: default and cure, where the collateral value exceeds the outstanding loan amount. We propose a three-step selection approach with a joint probability framework for default, cure (i.e., zero-LGD) and non-zero loss severity information. The proposed methodology demonstrates improved performance in out-of-time predictions compared to widely used OLS regressions. (C) 2018 Elsevier B.V. All rights reserved. article 2018 0377-2217,1872-6860 10.1016/j.ejor.2018.02.057 European Journal of Operational Research 270 ELSEVIER SCIENCE BV

AMSTERDAM

246-259 1 https://epub.uni-regensburg.de/id/eprint/37374 Hung Do Daniel Rösch Harald Scheule Banks are obliged to provide downturn estimates for loss given defaults (LGDs) in the internal ratings-based approach. While downturn conditions are characterized by systematically higher LGDs, it is unclear which factors may best capture these conditions. As LGDs depend on recovery payments which are collected during varying economic conditions in the resolution process, it is challenging to identify suitable economic variables. Using a Bayesian Finite Mixture Model, we adapt random effects to measure economic conditions and to generate downturn estimates. We find that systematic effects vary among regions, i.e., the US and Europe, and strongly deviate from the economic cycle. Our approach offers straightforward supportive tools for decision makers. Risk managers are enabled to select their individual margin of conservatism based on their portfolios, while regulators might set a lower bound to guarantee conservatism. In comparison to other approaches, our proposal appears to be conservative enough during downturn conditions and inhibits over-conservatism. (C) 2018 Elsevier B.V. All rights reserved. article 2018 0377-2217,1872-6860 10.1016/j.ejor.2018.05.059 European Journal of Operational Research 271 ELSEVIER SCIENCE BV

AMSTERDAM

1113-1144 https://epub.uni-regensburg.de/id/eprint/40080 Jennifer Betz Ralf Kellner Daniel Rösch This paper shows that the revised loan loss provisioning based on the International Financial Reporting Standards (IFRS) and the US Generally Accepted Accounting Principles (GAAP) implies a reduction of Tier 1 capital. The paper finds in a counterfactual analysis that these changes are more severe (i) during economic downturns, (ii) for credit portfolios of low quality, (iii) for banks that do not tighten capital standards during downturns, and (iv) under a more comprehensive definition of significant increase in credit risk (SICR) under IFRS. The provisioning rules further increase the procyclicality of bank capital requirements. Adjustments of the SICR threshold or capital buffers are suggested as ways to mitigate a regulatory pressure that may emerges due to the reduction of regulatory capital. (C) 2018 Elsevier B.V. All rights reserved. article 2018 1572-3089,1878-0962 10.1016/j.jfs.2018.02.009 Journal of Financial Stability 36 Elsevier

NEW YORK

114-129 https://epub.uni-regensburg.de/id/eprint/37263 Steffen Krüger Daniel Rösch Harald Scheule Credit risk analytics in R will enable you to build credit risk models from start to finish. Accessing real credit data via the accompanying website www.creditriskanalytics.net, you will master a wide range of applications, including building your own PD, LGD and EAD models as well as mastering industry challenges such as reject inference, low default portfolio risk modeling, model validation and stress testing. This book has been written as a companion to Baesens, B., Roesch, D. and Scheule, H., 2016. Credit Risk Analytics: Measurement Techniques, Applications, and Examples in SAS. John Wiley & Sons. book 2017 11 23 978-1977760869 Create Space Independent Publishing Platform https://epub.uni-regensburg.de/id/eprint/36395 Harald Scheule Daniel Rösch Bart Baesens This cumulative thesis contributes to the literature on credit risk modeling and focuses on comovements of risk parameters that intensify losses during recessions. The models provide more precise estimates of credit risk and a better understanding of systematic risk. This can improve risk-based capital reserves and can help to avoid a severe underestimation of risk and capital shortfalls in economic downturn periods. Furthermore, the discussion of regulatory requirements and the supervision of internal risk models can benefit from empirical results. The first study extends the scope of loss given default (LGD) modeling by proposing the quantile regression to separately regress each quantile of the distribution. This approach enables a new look on covariate and particularly downturn effects that vary over quantiles. The second study analyzes the length of workout processes by a Cox proportional hazards model. Systematic effects are examined by the inclusion of time-varying frailties. The third study presents a copula model for the lifetime expected loss that combines accelerated failure time models for the default time with a beta regression of the LGD. The use of copulas provide continuous-time LGD forecasts and flexible dependence structures between default risk and loss severity. The fourth study combines a Probit model for the probability of default and a fractional response model for the LGD to demonstrate the impact of revised loan loss provisioning on bank capital requirements. In addition, goodness-of-fit measures enable to validate these approaches. Simulation studies and analyses of representative portfolios provide implications and demonstrate the significance of empirical results. thesis_rgbg 2017 9 06 https://epub.uni-regensburg.de/id/eprint/36145 Steffen Krüger Literature on Losses Given Default (LGD) usually focuses on mean predictions, even though losses are extremely skewed and bimodal. This paper proposes a Quantile Regression (QR) approach to get a comprehensive view on the entire probability distribution of losses. The method allows new insights on covariate effects over the whole LGD spectrum. In particular, middle quantiles are explainable by observable covariates while tail events, e.g., extremely high LGDs, seem to be rather driven by unobservable random events. A comparison of the QR approach with several alternatives from recent literature reveals advantages when evaluating downturn and unexpected credit losses. In addition, we identify limitations of classical mean prediction comparisons and propose alternative goodness of fit measures for the validation of forecasts for the entire LGD distribution. (C) 2017 Elsevier B.V. All rights reserved. article 2017 6 0378-4266,1872-6372 10.1016/j.jbankfin.2017.03.001 Journal of Banking & Finance 79 ELSEVIER SCIENCE BV

AMSTERDAM

42-56 https://epub.uni-regensburg.de/id/eprint/35374 Steffen Krüger Daniel Rösch book 2017 978-1540622594 Create Space Independent Publishing Platform https://epub.uni-regensburg.de/id/eprint/36236 Daniel Rösch book 2016 10 978-1-119-14398-7 John Wiley & Sons

New York, N.Y.

https://epub.uni-regensburg.de/id/eprint/34707 Harald Scheule Bart Baesens Daniel Rösch Using a unique data base of Global Credit Data with individual loan information from small and medium sized entities in Germany, Great Britain and the United States, we evaluate the time to resolution of defaulted loans. A comparison across countries reveals country specific drivers for the resolution time which can be explained fairly well by differences in the regulatory and legal framework. Lenders seem to be aware of these differences and adjust their lending behavior in the limits set by these bankruptcy systems of the countries. (C) 2016 Elsevier Inc. All rights reserved. article 2016 8 1544-6123,1544-6131 10.1016/j.frl.2016.03.013 Finance Research Letters 18 ACADEMIC PRESS INC ELSEVIER SCIENCE

SAN DIEGO

7-31 https://epub.uni-regensburg.de/id/eprint/34545 Jennifer Betz Ralf Kellner Daniel Rösch The Basel Committee on Banking Supervision recently proposed fundamental changes in the regulatory treatment of financial institutions' trading book positions. Among others, a replacement of Value-at-Risk (alpha = 0.99) by Expected Shortfall (alpha = 0.975) for the quantification of market risk is recommended. While this increases capital requirements for heavy tailed risks, its consequences for model risk related to the estimation process have not been explored. Hence, the aim of this paper is to analyze how both risk measures react to different sources of model risk in order to better understand the impact of the intended change in risk measures. Our results show that the Expected Shortfall (alpha = 0.975) is more sensitive towards regulatory arbitrage and parameter misspecification. We find that this is based on a trade-off between a model's ability to better capture the heavy tailed behavior of risks and a higher vulnerability to model risk. These new aspects should be taken into account in the regulatory decision for Expected Shortfall (alpha = 0.975). (C) 2016 Elsevier B.V. All rights reserved. article 2016 5 0165-1889,1879-1743 10.1016/j.jedc.2016.05.002 Journal of Economic Dynamics and Control 68 ELSEVIER SCIENCE BV

AMSTERDAM

45-63 https://epub.uni-regensburg.de/id/eprint/34554 Ralf Kellner Daniel Rösch This paper explores whether factor based credit portfolio risk models are able to predict losses in severe economic downturns such as the recent Global Financial Crisis (GFC) within standard confidence levels. The paper analyzes (i) the accuracy of default rate forecasts, and (ii) whether forecast downturn percentiles (Value-at-Risk, VaR) are sufficient to cover default rate outcomes over a quarterly and an annual forecast horizon. Uninformative maximum likelihood and informative Bayesian techniques are compared as they imply different degrees of uncertainty. We find that quarterly VaR estimates are generally sufficient but annual VaR estimates may be insufficient during economic downturns. In addition, the paper develops and analyzes models based on auto-regressive adjustments of scores, which provide a higher forecast accuracy. The consideration of parameter uncertainty and auto-regressive error terms mitigates the shortfall. (C) 2015 Elsevier B.V. and Association of European Operational Research Societies (EURO) within the International Federation of Operational Research Societies (IFORS). All rights reserved. article 2016 0377-2217,1872-6860 10.1016/j.ejor.2015.09.007 European Journal of Operational Research 249 ELSEVIER SCIENCE BV

AMSTERDAM

440-456 https://epub.uni-regensburg.de/id/eprint/35393 Yongwoong Lee Daniel Rösch Harald Scheule This article analyzes the sensitivity to systematic credit risk and pricing in fixed income instruments and compares corporate bonds and asset securitizations. The article finds cross-sectional variation of systematic credit risk given the same credit rating and a market premium for the systematic risk embedded in yield spreads. Therefore, credit ratings do not provide comprehensive information on the degree of systematic risk, and investors are compensated for such differences in systematic risk after controlling for credit ratings and other risk characteristics. article 2016 10.3905/jfi.2016.26.1.042 Journal of Fixed Income 26 Institutional Investor Journals 42-60 https://epub.uni-regensburg.de/id/eprint/35391 Daniel Rösch Harald Scheule This paper analyses the systemic risk in relation to bank lending for Asian economies. The methodology complements existing market-based systemic risk measures by providing measures based on accounting information that regulators typically collect. Loan loss provisions of banks are decomposed into (i) a prediction component that is based on observable bank characteristics, and (ii) two frailty components: a bank-specific systematic factor based on the assumption that a bank's asset portfolio is diversified and a systemic factor. Systemic risk is measured as the Value-at-Risk and Expected Shortfall of the financial system based on a simulation model that takes into account the current condition of banks in the financial system, the absolute size and the capitalisation of financial institutions, as well as the sensitivity to systematic and systemic frailty risk. (C) 2016 Elsevier B.V. All rights reserved. article 2016 0927-538X,1879-0585 10.1016/j.pacfin.2016.01.002 Pacific-Basin Finance Journal 40 B ELSEVIER SCIENCE BV

AMSTERDAM

289-305 https://epub.uni-regensburg.de/id/eprint/35392 Daniel Rösch Harald Scheule In the recent literature, methods from extreme value theory (EVT) have frequently been applied to the estimation of tail risk measures. While previous analyses show that EVT methods often lead to accurate estimates for risk measures, a potential drawback lies in large standard errors of the point estimates in these methods, as only a fraction of the data set is used. Thus, we comprehensively study the impact of model risk on EVT methods when determining the value-at-risk and expected shortfall. We distinguish between first-order effects of model risk, which consist of misspecification and estimation risk, and second-order effects of model risk, which refer to the dispersion of risk measure estimates, and show that EVT methods are less prone to first-order effects. However, they show a greater sensitivity toward second-order effects. We find that this can lead to severe value-at-risk and expected shortfall underestimations and should be reflected in regulatory capital models. article 2016 1465-1211,1755-2842 10.21314/JOR.2016.337 Journal of Risk 18 INCISIVE MEDIA

LONDON

39-70 6 https://epub.uni-regensburg.de/id/eprint/34543 Harald Scheule Ralf Kellner Daniel Rösch We analyze the pricing of systematic risk factors in credit default swap (CDS) contracts in a two-stage empirical framework. Firstly we estimate contract-specific sensitivities (betas) to several systematic risk factors by time-series regressions using quoted CDS spreads of 339 U.S. entities from January 2004 to December 2010. Secondly, we show that these contract-specific sensitivities are cross-sectionally priced in CDS spreads after controlling for individual risk factors. We find that the credit market climate, the Cross-market Correlation, and the market volatility explain CDS spread changes and that their corresponding sensitivities (betas) are particularly priced in the cross-section. Our basic risk factors explain about 83% (90%) of the CDS spreads prior to (during) the crisis. article 2016 10.1016/j.qref.2016.06.007 Quarterly Review of Economics and Finance 64 Elsevier 183-195 https://epub.uni-regensburg.de/id/eprint/36235 Arndt Claussen Sebastian Löhr Daniel Rösch Harald Scheule This paper introduces a simple, non-parametric way of inferring risk-neutral credit stress event intensities for idiosyncratic, sectoral, and global shocks contained in market credit spreads. We provide an econometric analysis of the implied latent stress event dynamics. A vector autoregressive regression model with exogenous variables finds that these intensities can be related to an observable stock market index, the market volatility, the volatility skew, and treasury yields. (c) 2014 Wiley Periodicals, Inc. Jrl Fut Mark 35:300-320, 2015 article 2015 4 0270-7314,1096-9934 10.1002/fut.21695 Journal of Futures Markets 35 WILEY-BLACKWELL

HOBOKEN

300-320 4 https://epub.uni-regensburg.de/id/eprint/31537 Rainer Jobst Daniel Rösch Harald Scheule Martin Schmelzle The global financial crisis has shown that many financial institutions dealing with credit derivatives were exposed to severe unexpected losses. This indicates that systematic influences are decisively underestimated particularly with regard to structured products like securitized tranches of collateralized debt obligations. Our analytical study addresses these systematic effects: We provide a simple model which allows a closed-form comparison of both bonds and tranches with respect to their systematic risk. We demonstrate that the exposure to systematic risk of tranches may be many times higher than the exposure of bonds, even if both products share the same rating grade, e.g., an ‘AAA’ rating, measured by either default probability or expected loss. Particularly in economic downturns, default rates of tranches may be multiples of those of bonds. Our results help understand high default rates of tranches during the financial crisis and show that classical ratings are insufficient metrics for measuring risks of structured products. article 2014 1380-6645,1573-7144 10.1007/s11147-013-9089-1 Review of Derivatives Research 17 Springer 1-37 1 https://epub.uni-regensburg.de/id/eprint/28303 Arndt Claussen Sebastian Löhr Daniel Rösch This paper analyzes the level and cyclicality of regulatory bank capital for asset portfolio securitizations in relation to the cyclicality of capital requirements for the underlying loan portfolio as under Basel We find that the cyclicality of capital requirements is higher for (i) asset portfolio securitizations relative to primary loan portfolios, (ii) Ratings Based Approach (RBA) relative to the Supervisory Formula Approach, (iii) given the RBA for a point-in-time rating methodology relative to a rate-and-forget rating methodology, and (iv) under the passive reinvestment rule relative to alternative rules. Capital requirements of the individual tranches reveal that the volatility of aggregated capital charges for the securitized portfolio is triggered by the most senior tranches. This is due to the fact that senior tranches are more sensitive to the macroeconomy. An empirical analysis provides evidence that current credit ratings are time-constant and that economic losses for securitizations have exceeded the required capital in the recent financial crisis. (C) 2014 Elsevier B.V. All rights reserved. article 2014 0377-2217,1872-6860 10.1016/j.ejor.2014.01.011 European Journal of Operational Research 237 ELSEVIER SCIENCE BV

AMSTERDAM

289-302 1 https://epub.uni-regensburg.de/id/eprint/29757 Kristina Lützenkirchen Daniel Rösch Harald Scheule This paper evaluates the resurrection event regarding defaulted firms and incorporates observable cure events in the default prediction of SME. Due to the additional cure-related observable data, a completely new information set is applied to predict individual default and cure events. This is a new approach in credit risk that, to our knowledge, has not been followed yet. Different firm-specific and macroeconomic default and cure-event-influencing risk drivers are identified. The significant variables allow a firm-specific default risk evaluation combined with an individual risk reducing cure probability. The identification and incorporation of cure-relevant factors in the default risk framework enable lenders to support the complete resurrection of a firm in the case of its default and hence reduce the default risk itself. The estimations are developed with a database that contains 5930 mostly small and medium-sized German firms and a total of more than 23000 financial statements over a time horizon from January 2002 to December 2007. Due to the significant influence on the default risk probability as well as the bank's possible profit prospects concerning a cured firm, it seems essential for risk management to incorporate the additional cure information into credit risk evaluation. (C) 2014 Elsevier B.V. All rights reserved. article 2014 0377-2217,1872-6860 10.1016/j.ejor.2014.04.046 European Journal of Operational Research 238 ELSEVIER SCIENCE BV

AMSTERDAM

846-857 3 https://epub.uni-regensburg.de/id/eprint/31886 Daniel Rösch Marcus Wolter The global financial crisis exposed financial institutions to severe unexpected losses in relation to mortgage securitizations and derivatives. This article finds that risk models such as ratings are exposed to a large degree of systematic risk and parameter uncertainty. An out-of-sample forecasting exercise of the financial crisis shows that a simple approach addressing both issues is able to produce ranges for risk measures consistent with realized losses. This explains how financial markets were taken by surprise in relation to realized losses. article 2014 0022-4367,1539-6975 10.1111/j.1539-6975.2013.12009.x Journal of Risk and Insurance 81 WILEY

HOBOKEN

563-586 3 https://epub.uni-regensburg.de/id/eprint/61210 Daniel Rösch Harald Scheule This paper offers a joint estimation approach for forecasting probabilities of default and loss rates given default in the presence of selection. The approach accommodates fixed and random risk factors. An empirical analysis identifies bond ratings, borrower characteristics and macroeconomic information as important risk factors. A portfolio-level analysis finds evidence that common risk measurement approaches may underestimate bank capital by up to 17% relative to the presented model. article 2014 10.1057/jors.2012.82 Journal of the Operational Research Society 65 Palgrave Macmillan 393-407 3 https://epub.uni-regensburg.de/id/eprint/28306 Daniel Rösch Harald Scheule book 2013 978-1-11-996396-7 John Wiley & Sons

Chichester

Daniel Rösch und Harald Scheule https://epub.uni-regensburg.de/id/eprint/28367 Daniel Rösch Harald Scheule book 2013 978-111-996-396-7 (print), 978-111-996-604-3 (online) Wiley

Chichester

Daniel Rösch und Harald Scheule https://epub.uni-regensburg.de/id/eprint/29161 article 2013 Journal of Futures Markets 33 John Wiley & Sons 994-1023 11 https://epub.uni-regensburg.de/id/eprint/28305 Sebastian Löhr Olga Mursajew Daniel Rösch Harald Scheule Correlations are the main drivers for credit portfolio risk and constitute a major element in pricing credit derivatives such as synthetic single-tranche collateralized debt obligation swaps. This study suggests a dynamic panel regression approach to model and forecast implied correlations. Random effects are introduced to account for unobservable time-specific effects on implied tranche correlations. The implied-correlation forecasts of tranche spreads are compared to forecasts using historical correlations from asset returns. The empirical findings support our proposed dynamic mixed-effects regression correlation model. (c) 2013 Wiley Periodicals, Inc. article 2013 0270-7314,1096-9934 10.1002/fut.21626 Journal of Futures Markets 33 WILEY

HOBOKEN

994-1023 11 https://epub.uni-regensburg.de/id/eprint/62143 Sebastian Löhr Olga Mursajew Daniel Rösch Harald Scheule article 2013 Journal of Risk and Insurance 81 Blackwell Publishing 563-586 3 https://epub.uni-regensburg.de/id/eprint/28296 Daniel Rösch Harald Scheule This paper develops a framework to measure the exposure to systematic risk for pools of asset securitizations and measures empirically whether current ratings-based rules for regulatory capital of securitizations under Basel II and Basel III reflect this exposure. The analysis is based on a comprehensive US dataset on asset securitizations for the time period between 2000 and 2008. We find that the shortfall of regulatory capital during the Global Financial Crisis is strongly related to ratings. In particular, we empirically show that insufficient capital is allocated to tranches with the highest rating. These tranches account for the greatest part of the total issuance volumes. Furthermore, this paper is the first to calibrate risk weights which account for systematic risk and provide sufficient capital buffers to cover the exposure during similar economic downturns. These policy-relevant findings suggest a re-calibration of RBA risk weights and may contribute to the current efforts by the Basel Committee on Banking Supervision and others to re-establish sustainable securitization markets and to improve the stability of the financial system. article 2013 10.1016/j.jbankfin.2013.04.021 Journal of Banking and Finance 37 Elsevier 5236-5247 12 https://epub.uni-regensburg.de/id/eprint/28302 Kristina Lützenkirchen Daniel Rösch Harald Scheule The Global Financial Crisis (GFC) has led to a general discussion of the accuracy and declining standards of credit rating agency ratings. Substantial criticism has been directed toward the securitisation market, which has been identified as one of the main sources of the crisis. This study focuses on the ability of rating agencies to adjust their ratings prior to impairments of structured finance transactions. We develop a new measure that quantifies a rating agency’s performance in advance of defaults. Analysing a large number of impaired transactions rated by Moody’s Investors Service, we find that rating quality deteriorated during the GFC. Furthermore, we identify tranchespecific and macroeconomic factors that explain differences in Moody’s performance. article 2013 1351-847X,1466-4364 10.1080/1351847X.2011.636831 European Journal of Finance 19 Taylor & Francis 841-860 9 https://epub.uni-regensburg.de/id/eprint/28307 Matthias Bodenstedt Daniel Rösch Harald Scheule article 2012 10.1016/j.jbankfin.2011.02.026 Journal of Banking and Finance 36 Elsevier 733-748 3 https://epub.uni-regensburg.de/id/eprint/28308 Daniel Rösch Harald Scheule Dieser Artikel analysiert einen umfangreichen Datensatz mit Jahresabschluss und Ausfallinformationen deutscher, mittelständischer Unternehmen. Diese Daten, welche als typisch für ein Firmenkreditportfolio einer Großbank zu sehen sind, werden als Basis genutzt, um ein firmenspezifisches Verlustprognosemodell zu entwickeln. Unter Verwendung dieses Modells können signifikante firmenspezifische und makroökonomische Risikotreiber identifiziert und Ausfallrisiken über einen Mehrjahreshorizont prognostiziert werden. Über das zeitspezifische Verhalten der ermittelten Ausfallwahrscheinlichkeiten werden mehrperiodige Portfolioverlustverteilungen für bankspezifische Kreditportfolien geschätzt. Die Analysen basieren auf einem Datensatz aus 5.930 deutschen, mittelständischen Unternehmen. Zu diesen Unternehmen werden über einen Zeitraum von 2002 bis 2007 insgesamt über 23.000 Jahresabschlüsse analysiert. Die Ergebnisse können als Grundlage zur Entwicklung von Handlungsstrategien dienen, um Kreditportfolioverluste über mehrere Perioden realistischer bewerten zu können. This article analyses a comprehensive set of data including annual financial statements and default probability information relating to German small and medium-sized enterprises. This data set, which must be deemed typical of a business-customer credit portfolio of a large bank, is used as basis for developing an enterprise-specific default probability forecasting model. This model permits to identify significant company-specific and macroeconomic risk drivers and to forecast default probability risks over a multi-annual horizon. On the basis of the time-specific modes of behaviour of the default probabilities so ascertained, multi-period portfolio loss distributions have been estimated for bank-specific credit portfolios. The analyses are based on a data set relating to 5,930 German small and mediumsized enterprises. For these enterprises, a total of over 23,000 annual financial statements relating to the period 2002/2007 have been analysed. The results may be used as a basis on which to develop action strategies allowing credit portfolio losses to be more realistically estimated for several periods. article 2012 0023-4591,1865-5734 10.3790/kuk.45.2.189 Kredit und Kapital 45 Duncker und Humblot 189-217 2 https://epub.uni-regensburg.de/id/eprint/28309 Daniel Rösch Marcus Wolter article 2011 Risikomanager 9 Bank-Verlag Medien 20-21 https://epub.uni-regensburg.de/id/eprint/28341 Arndt Claussen Sebastian Löhr Kristina Lützenkirchen Daniel Rösch Harald Scheule This paper provides evidence for the relationship between credit quality, recovery rate, and correlation. The paper finds that rating grade, rating shift, and macroeconomic factors provide a highly significant explanation for default risk and recovery risk of US bond issues. The empirical data suggest that default and recovery processes are highly correlated. Therefore, a joint approach is required for estimating time-varying default probabilities and recovery rates that are conditional on default. This paper develops and applies such a model. article 2011 1354-7798,1468-036X 10.1111/j.1468-036X.2010.00582.x European Financial Management 17 Wiley-Blackwell 120-144 1 https://epub.uni-regensburg.de/id/eprint/28311 Benjamin Bade Daniel Rösch Harald Scheule article 2011 Journal of Risk Model Validation 5 Incisive Media 25-44 2 https://epub.uni-regensburg.de/id/eprint/28310 Benjamin Bade Daniel Rösch Harald Scheule article 2011 Journal of Risk Model Validation 5 Risk Journals; Incisive Media (London) 3-24 2 https://epub.uni-regensburg.de/id/eprint/21924 Alfred Hamerle Andreas Dartsch Rainer Jobst Kilian Plank article 2011 Risiko Manager 1 Bank-Verlag 1,8-17 https://epub.uni-regensburg.de/id/eprint/19124 Alfred Hamerle Rainer Jobst This paper provides an empirical study, which assesses the historical performance of credit rating agency (CRA) ratings for securitizations before and during the financial crisis. The paper finds that CRAs do not sufficiently address the systematic risk of the underlying collateral pools as well as characteristics of the deal and tranche structure in their ratings. The paper also finds that impairment risk is understated during origination years and years with high securitization volumes when CRA fee revenue is high. The mismatch between credit ratings of securitizations and their underlying risks has been suggested as one source of the Global Financial Crisis, which resulted in the criticism of models and techniques applied by CRAs and misaligned incentives due to the fees paid by originators. monograph 2011 58 https://epub.uni-regensburg.de/id/eprint/28342 Daniel Rösch Harald Scheule book_section 2010 978-0-470-05756-8 (gesamt) Encyclopedia of Quantitative Finance. Bd. 1 Wiley

Chichester

und https://epub.uni-regensburg.de/id/eprint/28313 Daniel Rösch This paper analyzes the level and cyclicality of bank capital requirement in relation to (i) the model methodologies through-the-cycle and point-in-time, (ii) four distinct downturn loss rate given default concepts, and (iii) US corporate and mortgage loans. The major finding is that less accurate models may lead to a lower bank capital requirement for real estate loans. In other words, the current capital regulations may not support the development of credit portfolio risk measurement models as these would lead to higher capital requirements and hence lower lending volumes. The finding explains why risk measurement techniques in real estate lending may be less developed than in other credit risk instruments. In addition, various policy recommendations for prudential regulators are made. article 2010 10.1111/j.1468-2443.2009.01102.x International Review of Finance 10 Wiley-Blackwell 185-207 2 https://epub.uni-regensburg.de/id/eprint/28312 Daniel Rösch Harald Scheule book_section 2010 1906348251, 9781906348250 Model Risk – Identification, Measurement and Management Risk Books

London

Daniel Rösch und Harald Scheule https://epub.uni-regensburg.de/id/eprint/28345 Daniel Rösch Harald Scheule book_section 2010 1906348383, 9781906348380 Reinventing Retail Lending Analytics Risk Books

London

und https://epub.uni-regensburg.de/id/eprint/28351 Daniel Rösch Harald Scheule book 2010 1906348251, 9781906348250 Risk Books

London

Daniel Rösch und Harald Scheule https://epub.uni-regensburg.de/id/eprint/28368 Daniel Rösch Harald Scheule article 2010 Uni-Magazin, Leibniz Universität Hannover https://epub.uni-regensburg.de/id/eprint/28344 Michael Breitner Daniel Rösch Grigoriy Tymchenko Hans-Jörg von Mettenheim article 2010 Sparkassenzeitschrift 73 46 https://epub.uni-regensburg.de/id/eprint/28343 Daniel Rösch Recent studies find a positive correlation between default and loss given default rates of credit portfolios. In response, financial regulators require financial institutions to base their capital on ‘Downturn’ loss rates given default which are also known as Downturn LGDs. This article proposes a concept for the Downturn LGD which incorporates econometric properties of credit risk as well as the information content of default and loss given default models. The concept is compared to an alternative proposal by the Department of the Treasury, the Federal Reserve System and the Federal Insurance Corporation. An empirical analysis is provided for US American corporate bond portfolios of different credit quality, seniority and security. article 2009 0963-8008,1468-0416 10.1111/j.1468-0416.2008.00145.x Financial Markets, Institutions and Instruments 18 Wiley-Blackwell 1-26 1 https://epub.uni-regensburg.de/id/eprint/28314 Daniel Rösch Harald Scheule article 2009 Journal of Risk Model Validation 2 Incisive Media 3-11 4 https://epub.uni-regensburg.de/id/eprint/28315 Daniel Rösch Harald Scheule article 2009 OR News 74-75 36 https://epub.uni-regensburg.de/id/eprint/28347 Michael Breitner Daniel Rösch Hans-Jörg von Mettenheim other 2009 Daniel Rösch und Harald Scheule https://epub.uni-regensburg.de/id/eprint/28379 Daniel Rösch Harald Scheule article 2008 8 01 Risk 21 Risk Waters Group 78-82 https://epub.uni-regensburg.de/id/eprint/8004 Daniel Rösch Birker Winterfeldt article 2008 Risiko-Manager 22 Bank-Verlag 1,8-14 https://epub.uni-regensburg.de/id/eprint/8003 Alfred Hamerle Rainer Jobst Hans-Jochen Schropp monograph 2008 15 https://epub.uni-regensburg.de/id/eprint/28346 Daniel Rösch Harald Scheule One of the most significant developments in international credit markets in recent years has been the trade in Collateralized Debt Obligations (CDO), which has enabled financial institutions to repackage the credit risk of an asset portfolio into tranches to be transferred to investors. The present paper evaluates the credit risk of such a portfolio and the related tranches by applying two prominent prototypes for credit ratings, namely the point-in-time and through-the-cycle approach. The central parameters default probability and correlation are forecast for multiple years and related forecasting errors are included. The article's main findings are that banks which transfer debt tranches but retain an equity part and apply a through-the-cycle rating approach may be exposed to higher insolvency risk. Firstly, the credit risk retained may be underestimated resulting in an inadequate capital allocation. Secondly, the credit risk transferred may be overestimated resulting in additional risk-based transfer costs. article 2008 10.1016/j.gfj.2008.09.007 Global Finance Journal 19 Elsevier 235-251 3 https://epub.uni-regensburg.de/id/eprint/28316 Daniel Rösch Harald Scheule book_section 2008 978-1-906348-11-3 ; 1-906348-11-1 Stress-testing for Financial Institutions - Applications, Regulations, and Techniques Risk Books Daniel Rösch und Harald Scheule 3-16 https://epub.uni-regensburg.de/id/eprint/28349 Daniel Rösch Harald Scheule article 2008 Risiko-Manager 9 Bank-Verlag 1,8-15 https://epub.uni-regensburg.de/id/eprint/8006 Alfred Hamerle Rainer Jobst Matthias Lerner book 2008 978-3-86553-260-2 WiKu-Verl.

Duisburg

https://epub.uni-regensburg.de/id/eprint/8377 Rainer Jobst book_section 2008 978-1-906348-11-3 Stress Testing for Financial Institutions: Applications, Regulations and Techniques Riskbooks

London

Daniel Rösch und Harald Scheule 67-91 https://epub.uni-regensburg.de/id/eprint/8000 Alfred Hamerle Rainer Jobst Michael Knapp Matthias Lerner book 2008 978-1-906348-11-3 ; 1-906348-11-1 Risk Books

London

Daniel Rösch und Harald Scheule https://epub.uni-regensburg.de/id/eprint/28370 Daniel Rösch Harald Scheule article 2007 Journal of Credit Risk 3 Incisive Media 113-134 4 https://epub.uni-regensburg.de/id/eprint/28318 Daniel Rösch Harald Scheule article 2007 Journal of Financial Forecasting 1 Risk Journals 25-54 2 https://epub.uni-regensburg.de/id/eprint/8204 Alfred Hamerle Rainer Jobst Thilo Liebig Daniel Rösch article 2007 Journal of Risk Model Validation 1 Incisive Media 55-75 1 https://epub.uni-regensburg.de/id/eprint/28320 Daniel Rösch Harald Scheule book_section 2006 3-540-33085-2; 978-3-540-33085-1 The Basel II Risk Parameters Springer

Berlin

Bernd Engelmann und Robert Rauhmeier 105-126 https://epub.uni-regensburg.de/id/eprint/8211 Daniel Rösch Harald Scheule book_section 2006 3-8006-3289-6 Wirtschaftsstatistik: Festschrift zum 65. Geburtstag von Professor Dr. Dr. h.c. mult. Eberhard Schaich Vahlen

München

Hans Wolfgang Brachinger Alfred Hamerle Ralf Münnich und Walter Schweitzer 65-79 https://epub.uni-regensburg.de/id/eprint/8209 Alfred Hamerle Daniel Rösch article 2006 Journal of Credit Risk 2 Incisive Media 101-122 4 https://epub.uni-regensburg.de/id/eprint/8206 Alfred Hamerle Daniel Rösch This article develops a simultaneous multifactor model for defaults and recoveries. Applying this model, risk parameters can be forecast using systematic and idiosyncratic risk factors and their implied correlations. The theoretical framework is accompanied by an empirical analysis in which a negative correlation between defaults and recoveries over the business cycle is observed. In the study, default and recovery rates are modeled by business cycle indicators, and the properties of the economic and regulatory capital given these risk drivers are shown. article 2005 10.3905/jfi.2005.591610 Journal of Fixed Income 15 Inst. Investor 63-75 2 https://epub.uni-regensburg.de/id/eprint/28322 Daniel Rösch Harald Scheule The New Basel Capital Accord will allow the determination of banks' regulatory capital requirements due to probabilities of default (PDs) which are estimated and forecasted from internal ratings. Broadly, two rating philosophies are distinguished: through the cycle versus point in time ratings. We employ a likelihood ratio backtesting of both types with respect to their probability of default forecasts and correlations derived from a nonlinear random effects panel model using data from Standard & Poor's. The implications for risk capital using these different philosophies are demonstrated. It is shown that Point in Time Ratings will exhibit much lower correlations and, thus, default probability forecasts should be more precise. As a consequence, Value-at-Risk quantiles of default distributions should be lower than those generated by Through the Cycle Ratings. Nevertheless, banks which use Point in Time Ratings may be punished in times of economic stress if the implied reduction of asset correlation is not taken into account. (C) 2004 International Institute of Forecasters. Published by Elsevier B.V. All rights reserved. article 2005 10.1016/j.ijforecast.2004.04.001 International Journal of Forecasting 21 ELSEVIER SCIENCE BV

AMSTERDAM

37-51 1 https://epub.uni-regensburg.de/id/eprint/8225 Daniel Rösch Im vorliegenden Beitrag wird untersucht, welche Auswirkungen Schätz- und Prognoserisiken auf das Backtesting von Ausfallwahrscheinlichkeiten aus Ratingsystemen haben. In der Regel wird unterstellt, dass alle Schuldner einer Ratingklasse dieselbe Ausfallwahrscheinlichkeit haben und die Richtigkeit einer Vorgabe dieser ex-ante unbekannten Ausfallwahrscheinlichkeit soll überprüft werden. Häufig wird für diese Vorgabe eine Schätzung aus historischen Ausfallzeitreihen herangezogen und in der Nullhypothese wird die Ausfallwahrscheinlichkeit gleich der Schätzung gesetzt («Schätzungen sind korrekt»). In diesem Fall bleibt jedoch unberücksichtigt, dass die Schätzungen Realisierungen von Zufallsgrössen und damit fehlerbehaftet sind. Als Folge wird das Ratingsystem einer Bank mit höherer Wahrscheinlichkeit als durch den Test suggeriert fälschlicherweise als mangelhaft eingestuft. article 2005 Die Unternehmung 59 Versus 535-546 6 https://epub.uni-regensburg.de/id/eprint/8217 Alfred Hamerle Daniel Rösch article 2005 Die Betriebswirtschaft 65 Schäffer-Poeschel 179-196 2 https://epub.uni-regensburg.de/id/eprint/5140 Alfred Hamerle Daniel Rösch In addition to “classical” approaches, such as the Gaussian CreditMetrics or Basel II model, the use of other copulas has recently been proposed in the area of credit risk for modeling loss distributions, particularly T copulas which lead to fatter tails ceteris paribus. As an amendment to recent research this paper shows some estimation results when the copula in a default-mode framework using a latent variable distribution is misspecified. It turns out that parameter estimates may be biased, but that the resulting forecast for the loss distribution may still be adequate. We also compare the performance of the true and misspeci fied models with respect to estimation risk. Finally, we demonstrate the ideas using rating agencies data and show a simple way of dealing with estimation risk in practice. Overall, our findings on the robustness of the Gaussian copula considerably reduce model risk in practical applications. article 2005 Journal of Risk 8 Incisive Media Plc 41-58 1 https://epub.uni-regensburg.de/id/eprint/8223 Alfred Hamerle Daniel Rösch article 2005 Credit Technology 53 35-42 https://epub.uni-regensburg.de/id/eprint/28358 Daniel Rösch Harald Scheule article 2005 Wilmott Magazine 2-6 https://epub.uni-regensburg.de/id/eprint/8214 Stefan Blochwitz Alfred Hamerle Stefan Hohl Robert Rauhmeier Daniel Rösch article 2005 Kredit und Rating Praxis 31 Rek & Thomas Medien AG 15-19 1 https://epub.uni-regensburg.de/id/eprint/8224 Alfred Hamerle Daniel Rösch article 2004 11 15 0341-4019,0341-6194,0340-8485 Zeitschrift für das gesamte Kreditwesen 57 Knapp 1275-1278 22 https://epub.uni-regensburg.de/id/eprint/8227 Stefan Blochwitz Alfred Hamerle Stefan Hohl Robert Rauhmeier Daniel Rösch thesis 2004 https://epub.uni-regensburg.de/id/eprint/8383 Daniel Rösch book_section 2004 3-540-20738-4 CreditRisk+ in the Banking Industry Springer

Berlin

Matthias Gundlach und Frank Lehrbaß 231-248 https://epub.uni-regensburg.de/id/eprint/28329 Leif Boegelein Alfred Hamerle Michael Knapp Daniel Rösch monograph 2004 https://epub.uni-regensburg.de/id/eprint/8228 Alfred Hamerle Rainer Jobst Christian Tegelkamp Markus Wadè article 2004 10.1108/eb022983 Journal of Risk Finance 5 Emerald 16-32 2 https://epub.uni-regensburg.de/id/eprint/8238 Daniel Rösch Harald Scheule article 2004 Kredit und Rating Praxis 30 Rek & Thomas Medien AG 20-22 6 https://epub.uni-regensburg.de/id/eprint/8237 Alfred Hamerle Daniel Rösch article 2004 1 Deutsches Risk 4 Incisive Financial Publ. 39-45 https://epub.uni-regensburg.de/id/eprint/8229 Alfred Hamerle Thilo Liebig Daniel Rösch article 2003 11 Risk 16 Incisive Financial Publ. 77-81 11 https://epub.uni-regensburg.de/id/eprint/8240 Alfred Hamerle Thilo Liebig Daniel Rösch article 2003 5 0341-2687,0036-6196 Schmalenbachs Zeitschrift für betriebswirtschaftliche Forschung (ZfbF) 55 Verl.-Gruppe Handelsblatt 199-223 https://epub.uni-regensburg.de/id/eprint/8243 Alfred Hamerle Daniel Rösch article 2003 1555-4961,1555-497X 10.1007/s11408-003-0303-2 Financial Markets and Portfolio Management 17 Swiss Society for Financial Market Research; Springer 309-331 3 https://epub.uni-regensburg.de/id/eprint/28332 Daniel Rösch monograph 2003 3–935821–70–0 2 Dt. Bundesbank

Frankfurt am Main

https://epub.uni-regensburg.de/id/eprint/8241 Alfred Hamerle Thilo Liebig Daniel Rösch article 2003 Risk Management Association Journal RMA 66-69 https://epub.uni-regensburg.de/id/eprint/8249 Daniel Rösch Harald Scheule article 2002 7 Die Bank Bank-Verlag 470-473 https://epub.uni-regensburg.de/id/eprint/8252 Alfred Hamerle Thilo Liebig Daniel Rösch article 2002 1435-246X,1210-0269,1335-1443 Central European Journal of Operations Research 10 Springer 163-186 https://epub.uni-regensburg.de/id/eprint/8255 Daniel Rösch article 2001 0172-7400,0342-4928 Kredit-Praxis 27 Gabler 8-13 1 https://epub.uni-regensburg.de/id/eprint/8364 Daniel Rösch In jüngerer Zeit werden in zunehmendem Maße Ansätze der Arbitrage Pricing Theory im praktischen Portfoliomanagement eingesetzt. Eine wichtige Klasse stellen die fundamentalen Faktoren-Modelle dar, bei denen unternehmensspezifische Variablen, wie z.B. Kurs/Gewinn-Verhältnis, Quotient aus Buch- und Marktwert, Dividendenrendite, Unternehmensgröße, historische Betas, als bewertungsrelevante Risikofaktoren vorab spezifiziert und in einem statistischen Querschnittsregressionsmodell empirisch auf Signifikanz geprüft werden. Eine andere Klasse von APT-Ansätzen spezifiziert die Faktoren durch makroökonomische Variablen, z.B. Inflationsrate, Zins oder Ölpreis. In einem ersten Schritt werden anhand von Zeitreihenregressionen die Sensitivitälen (Faktor-Betas) bezüglich der makroökonomischen Faktoren geschätzt, im zweiten Schritt wird die Querschnittsbeziehung zwischen Renditen und Sensitivitäten analysiert. Die zu den statistisch signifikanten Sensitivitäten gehörenden makroökonomischen Variablen werden als bewertungsrelevant angesehen. Im vorliegenden Beitrag wird gezeigt, daß eine derartige Vorgehensweise in aller Regel nicht gerechtfertigt ist und zu unzutreffenden Schlußfolgerungen in bezug auf die Bewertungsrelevanz der Risikofaktoren führen kann. Es zeigt sich, daß die den empirischen Tests zugrundeliegende Bewertungsgleichung im allgemeinen unsystematisches Risiko enthält. Als Folge davon sind sämtliche Schätzungen der Regressionskoeffizienten in der Querschnittsregression verzerrt. Damit sind die Beurteilung der Signifikanz der in den Ansatz aufgenommenen Variablen und der Versuch, die Faktoren empirisch zu identifizieren, nicht mehr möglich. Ferner wird gezeigt, daß auch in dem unrealistischen Fall einer exakten Faktorbewertung ohne unsystematisches Risiko die den Renditegenerierungsprozeß determinierenden Faktoren bekannt sein müssen. Dies steht jedoch exakt im Widerspruch zu den im Portfoliomanagement eingesetzten Ansätzen, die die bewertungsrelevanten Risikofaktoren auf empirischem Weg anhand eines statistischen Querschnittsmodells identifizieren möchten. article 1998 4 10.1007/BF01539864 OR Spectrum 20 Springer Verlag 123-134 2 https://epub.uni-regensburg.de/id/eprint/8374 Alfred Hamerle Daniel Rösch book 1998 3-8244-6729-1 Dt. Univ.-Verl.

Wiesbaden

https://epub.uni-regensburg.de/id/eprint/28372 Daniel Rösch article 1998 Die Betriebswirtschaft 58 Schäffer-Poeschel 38-48 1 https://epub.uni-regensburg.de/id/eprint/8376 Alfred Hamerle Daniel Rösch article 1997 Schmalenbachs Zeitschrift für betriebswirtschaftliche Forschung (ZfbF) 49 Verlagsgruppe Handelsblatt 858-876 10 https://epub.uni-regensburg.de/id/eprint/28337 Alfred Hamerle Daniel Rösch article 1996 0002-6018,1614-0176 Allgemeines Statistisches Archiv 80 Springer Verlag 361-370 4 https://epub.uni-regensburg.de/id/eprint/28338 Alfred Hamerle Daniel Rösch article 1996 Allgemeines Statistisches Archiv 80 Springer Verlag 299-312 3 https://epub.uni-regensburg.de/id/eprint/28339 Alfred Hamerle Daniel Rösch Many empirical studies have found little relation between sample mean returns of securities and estimated betas. Some of the studies have uncovered variables other than beta (e.g. firm size, ratio of book-to-market value, price/earnings ratio) that have power in explaining the sample cross-sectional variation in mean returns. The present paper shows that a possible explanation is that market portfolio proxies are mean-variance inefficient. It is shown that inefficiency of the market proxy renders estimates of the regression coefficients of the cross-sectional relationship biased. In order to illustrate the problems an artificial capital market is generated. In this financial market the CAPM is valid. It turns out that the bias of the parameter estimates may be considerable, and inference about the validity of the CAPM or about the significance of "anomalies" based on these estimates may be extremely misleading. article 1996 Financial Markets and Portfolio Management 10 Swiss Society for Financial Market Research; Springer 61-74 1 https://epub.uni-regensburg.de/id/eprint/28340 Alfred Hamerle Daniel Rösch