Risk factor contributions in portfolio credit risk models

In portfolio credit risk models, correlated credit events are often modelled by means of correlated latent variables. The latent variables are interpreted as the firms’ asset values, and assumed to follow a normal distribution. A procedure is described that uses the information embodied in rating transition matrices to infer the shape of the latent variable distribution. Applying the approach to transition matrices of different origin yields consistent results. Compared to the normal distribution, the implied asset value distributions are fat-tailed, leading to a substantial increase of portfolio credit risk relative to the one under normality. The results are thus highly relevant for the design of credit risk measurement systems.

Parameterizing Credit Risk Models

We consider the problem of calculating tail loss probability and conditional excess for the Bernoulli mixture model of credit risk. This is an important problem as all credit risk models proposed in literature can be represented as Bernoulli mixture models. Thus, we deviate from the efficient simulation of credit risk literature in that we propose an efficient simulation algorithm for this general Bernoulli mixture model in contrast to previous works that focus on specific credit risk models like CreditRisk\(^+\) or Credit Metrics. The algorithm we propose is a combination of stratification, importance sampling based on cross-entropy, and inner replications using the geometric shortcut method. We evaluate the efficiency of our general method considering three different examples: CreditRisk\(^+\) and two of the latent variable models, the Gaussian and the t-copula model. Numerical results suggest that the proposed general algorithm is more efficient than the benchmark methods for these specific models.

This chapter is a tutorial on portfolio risk management. It describes how to identify the primary sources of risk in complex portfolios—the “hot spots”—and how to identify the trades that will reduce those risks—the “hedges.” While this topic, identifying and reducing risk, would seem to be at the heart of risk management, many of the ideas presented here are not well-understood. Unfortunately, too much of the recent focus in risk management has been directed elsewhere—for example, toward under-standing and incorporating the risks of derivative securities. While derivtives can create complex and highly nonlinear payoffs, they are often not the primary source of risk in a portfolio. Also, while risk managers have focused much attention recently on developing firm wide systems to aggregate information from many traders or portfolio managers, they have generally not yet concentrated on how to use that information to obtain a comprehensive understanding of enterprise risk. More generally, while risk managers have also recently developed a greater appreciation of the statistical nature of risk and the need to use statistical measures such as value-at-risk (VaR)1 to quantify, monitor, and set limits on risk-taking activities, the tools required to understand and incorporate these statistical measures in portfolio risk management have lagged behind.2 This most important area of risk management—the development of portfolio analytics based on the statistical understanding of risk—has not received as much attention as it deserves. We hope this chapter helps to address this concern by describing a number of portfolio analysis tools that we have developed at Goldman Sachs in recent years to better understand, manage, and monitor the risks in our clients’ investment portfolios as well as our own trading positions. We focus on the concepts of risk management and how they apply in realistic trading and portfolio management contexts rather than on details of implementation.

The measurement of portfolio credit risk focuses on rare but significant large‐loss events. This paper investigates rare event asymptotics for the loss distribution in the widely used Gaussian copula model of portfolio credit risk. We establish logarithmic limits for the tail of the loss distribution in two limiting regimes. The first limit examines the tail of the loss distribution at increasingly high loss thresholds; the second limiting regime is based on letting the individual loss probabilities decrease toward zero. Both limits are also based on letting the size of the portfolio increase. Our analysis reveals a qualitative distinction between the two cases: in the rare‐default regime, the tail of the loss distribution decreases exponentially, but in the large‐threshold regime the decay is consistent with a power law. This indicates that the dependence between defaults imposed by the Gaussian copula is qualitatively different for portfolios of high‐quality and lower‐quality credits.

The Gaussian copula model, with all of its flaws, has become the most common approach for estimating the loss distribution of a portfolio of risky credits. In general, extensive calculations are still required, although simplification is often obtained by using an asymptotic single factor model. The model assumes idiosyncratic risks are fully diversified and dependence across names comes entirely from their joint dependence on a single common factor. However, credit risk exposure in a portfolio, such as one held by a bank, does not depend only on the probability of default, but also on uncertainty over loss given default (LGD) and even uncertainty over face amounts, or the exposure at default (EAD), in the case of revolving credits. This article extends the asymptotic model by formally introducing creditors’ draw rates and recoveries in case of default, both of which are tied to the same common factor. A closed-form asymptotic solution incorporating all three components of risk is obtained when distributions are continuous. The author also derives an approximate solution for the discrete distribution case and uses it to show how the overall loss distribution is affected by the properties of the LGD, in particular, its skewness.

The modeling of dependence structures (or copulas) is undoubtedly one of the key challenges for modern financial engineering. First applied to credit-risk modeling, copulas are now widely used across a range of derivatives transactions, asset pricing techniques, and risk models, and are a core part of the financial engineer's toolkit. However, by their very nature, copulas are complex and their applications are often misunderstood. Incorrectly applied, copulas can be hugely detrimental to a model or algorithm. Financial Engineering with Copulas Explained is a reader-friendly, yet rigorous introduction to the state-of-the-art regarding the theory of copulas, their simulation and estimation, and their use in financial applications. Starting with an introduction to the basic notions, such as required definitions and dependence measures, the book looks at statistical issues comprising parameter estimation and stochastic simulation. The book will show, from a financial engineering perspective, how copula theory can be applied in the context of portfolio credit-risk modeling, and how it can help to derive model-free bounds for relevant risk measures. The book will cover numerous different market applications of copulas, and enable readers to construct stable, high-dimensional models for asset pricing and risk modeling. Written to appeal to quantitatively minded practitioners across the trading floors and in risk management, academics and students, Financial Engineering with Copulas Explained will be a valuable, accessible and practical guide to this complex topic.

For the fund-of-funds manager, managing risk in a portfolio of hedge funds requires, first, recognizing that the return distribution of hedge funds is not normal and, second, taking into account the market crisis events that should happen rarely but that, in fact, occur every two to three years. Fortunately, value at risk methodology can be adapted for measuring the normal risk in the fund portfolio and, with some modification, can be applied to estimate how much the portfolio might lose in a crisis. In addition, modeling crises as coherent events allows one to estimate the likely amount of loss requiring a hedge, even though the particular nature of a future crisis may be unknown.

The Arbor City Community Foundation (ACCF) was a medium-sized endowment established in Illinois in the late 1970s through the hard work of several local families. The vision of the ACCF was to be a comprehensive center for philanthropy in the greater Arbor City region. ACCF had a fund balance (known collectively as “the fund”) of just under $240 million. The ACCF board of trustees had appointed a committee to oversee investment decisions relating to the foundation assets. The investment committee, under the guidance of the board, pursued an active risk-management policy for the fund. The committee members were primarily concerned with the volatility and distribution of portfolio returns. They relied on the value-at-risk (VaR) methodology as a measurement of the risk of both short- and mid-term investment losses. The questions in Part (A) of the case direct the students to analyze the risk inherent in both one particular asset and the entire ACCF portfolio. For this analysis the students need to calculate daily VaR and monthly VaR values and interpret these figures in the context of ACCF's risk management. In Part (B) the foundation receives a major donation. As a result, the risk inherent in its portfolio changes considerably. The students are asked to evaluate the risk of the fund's new portfolio and to perform a portfolio rebalancing analysis.Understanding the concept of value at risk (VaR); Calculating daily and monthly VaR by two different methods, the historical and the parametric approach; Interpreting the results of VaR calculations; Understanding the role of diversification for managing risk; Evaluating the impact of portfolio rebalancing on the overall risk of a portfolio.

This paper explores the use of a combined approach of Markowitz portfolio theory and Monte Carlo simulations for optimizing asset allocation and risk management in financial markets. First, we introduce the portfolio optimization framework of the Markowitz theory, focusing on how to construct an efficient portfolio by minimizing risk and maximizing expected return. Next, by integrating the Monte Carlo simulation method, the paper simulates asset returns and risk fluctuations under different market scenarios, aiming to assess the potential risks of various asset portfolios. To make the model more realistic, we assume an efficient market, rational investors, and neglect transaction costs and market frictions. Additionally, we consider the covariance matrix between assets, assuming its stability during the analysis period. Finally, based on a single-period investment model, the paper proposes an optimized asset allocation strategy, offering a more scientifically grounded decision-making approach for investors. The results show that combining these two methods can effectively improve the accuracy of asset allocation and reduce the overall risk of the portfolio. This study provides a new perspective and practical solution for financial risk assessment and asset management.

Risk management in translation service companies is an effective, integrative, proactive way to handle risk. Such companies must integrate defined long-term risk management into their strategies, decision-making, and daily processes; top-management must fully support the risk management system; risk factors, management benefits, and processes require articulation amongst employees; job roles, responsibilities, performance evaluation, and motivation systems must incorporate risk management practices. Risk management in project-based companies might also account for changing clients’ requirements, tight deadlines and budgets, different participants, and high IT use. However, this requires identifying the critical success factors. The research method builds upon the analysis and systemization of the scientific literature – from risk management, emphasizing the underlying strategic approach and effective management, to deriving the focal-points of effective risk management of translation projects. This work unifies the gathered knowledge and results in a conceptual model that integrates the specific assessment criteria for project risk management in translation companies – process definiteness and versatility, responsibility definiteness, top-management involvement, and risk management communication. The operationalisation of the model may lead to the companies’ critical success performance. To the best of our knowledge, this work is one of the first contributions addressing risk management assessment criteria in this industry.

Systematic risk has been a focus for stress testing and risk capital assessment. Under the Vasicek asymptotic single risk factor model framework, entity default risk for a risk homogeneous portfolio divides into two parts: systematic and entity specific. While entity specific risk can be modelled by a probit or logistic model using a relatively short period of portfolio historical data, modeling of systematic risk is more challenging. In practice, most default risk models do not fully or dynamically capture systematic risk. In this paper, we propose an approach to modeling systematic and entity specific risks by parts and then aggregating together analytically. Systematic risk is quantified and modelled by a multifactor Vasicek model with a latent residual, a factor accounting for default contagion and feedback effects. The asymptotic maximum likelihood approach for parameter estimation for this model is equivalent to least squares linear regression. Conditional entity PDs for scenario tests and through-the-cycle entity PD all have analytical solutions. For validation, we model the point-in-time entity PD for a commercial portfolio, and stress the portfolio default risk by shocking the systematic risk factors. Rating migration and portfolio loss are assessed.

This paper presents a comparative analysis of three foundational portfolio management models: the Markowitz Efficient Frontier, the Capital Asset Pricing Model (CAPM), and Multi-Factor Models. It aims to explore their theoretical foundations, empirical applicability, and practical usage to assess their strengths and limitations in financial portfolio management. The Markowitz model is appreciated for its detailed risk-return optimization framework but criticized for its reliance on historical data and computational demands. The CAPM simplifies risk assessment through market beta, yet its assumptions of market efficiency and a risk-free rate are viewed as impractical under current market conditions. Multi-Factor Models address some limitations of the CAPM by incorporating various risk factors, offering enhanced explanatory power but also adding complexity in factor selection.The study concludes that while no single model is universally superior, the choice of model should depend on the specific needs of the portfolio manager, including risk-return objectives and market environment. The paper suggests a blended or adaptive approach to model selection, highlighting the need for flexibility in response to evolving market dynamics and advances in computational finance.

Was Bank Risk Management Difficult in Turkey During the Banking Crisis There? Why Value at Risk Methods Won't Fix the Problems of Turkey's Banks

Internal ratings systems, implied credit risk and the consistency of banks’ risk classification policies