Counterparty Credit Risk Management Research Articles

Deep xVA Solver: A Neural Network–Based Counterparty Credit Risk Management Framework

In this paper, we present a novel computational framework for portfolio-wide risk management problems, where the presence of a potentially large number of risk factors makes traditional numerical techniques ineffective. The new method utilises a coupled system of BSDEs for the valuation adjustments (xVA) and solves these by a recursive application of a neural network based BSDE solver. This not only makes the computation of xVA for high-dimensional problems feasible, but also produces hedge ratios and dynamic risk measures for xVA, and allows simulations of the collateral account.

Boundary-safe PINNs extension: Application to non-linear parabolic PDEs in counterparty credit risk

The goal of this work is to develop a novel strategy for the treatment of the boundary conditions for multi-dimension nonlinear parabolic PDEs. The proposed methodology allows to get rid of the heuristic choice of the weights for the different addends that appear in the loss function related to the training process. It is based on defining the losses associated to the boundaries by means of the PDEs that arise from substituting the related conditions into the model equation itself. The approach is applied to challenging problems appearing in quantitative finance, namely, in counterparty credit risk management. Further, automatic differentiation is employed to obtain accurate approximation of the partial derivatives, the so called Greeks, that are very relevant quantities in the field.

Deep xVA Solver – A Neural Network Based Counterparty Credit Risk Management Framework

In this paper, we present a novel computational framework for portfolio-wide risk management problems, where the presence of a potentially large number of risk factors makes traditional numerical techniques ineffective. The new method utilises a coupled system of BSDEs for the valuation adjustments (xVA) and solves these by a recursive application of a neural network based BSDE solver. This not only makes the computation of xVA for high-dimensional problems feasible, but also produces hedge ratios and dynamic risk measures for xVA, and allows simulations of the collateral account.

Applying Machine Learning for Troubleshooting Credit Exposure and xVA Profiles

We explore the possibility of applying machine learning techniques to characterizing and troubleshooting problem in credit exposure profiles used in counterparty credit risk management and XVA. Machine learning models are trained to identify characteristics from the daily credit exposure profiles generated through Monte Carlo simulation and identify anomalies due to either counterparty behavioral changes or errors in inputs or calculation. In the paper, we have trained both K-means clustering and Convolutional Neural Network to a training data set containing more than 400,000 credit exposure profiles. The performance of models is satisfactory and we believe this is a promising area for further exploration. Identifying anomalies provides credit analysts early insight into emerging credit risk and errors in the data lineage.

Evaluation research on commercial bank counterparty credit risk management based on new intuitionistic fuzzy method

Over the past few years, the world economy is still in a profound adjustment stage after financial crisis, and it will continue to maintain a “New Mediocre” situation. It is clear that multiple risk factors, such as the issue of Brexit and European refugees, will increase the uncertainty of world economy growth. Under such economic development condition, it is obvious that the development of commercial bank will face a challenge; especially, the development of off-balance sheet business has received more attention from the commercial bank. Therefore, the counterparty credit risk has been brought into focus by the government and regulatory authority. This paper employs the new intuitionistic fuzzy method to improve the score function, and it aims to establish an evaluation mechanism of commercial bank counterparty credit risk management. By selecting the representative Chinese commercial banks, this paper conducts an empirical analysis to verify the validity of the evaluation system and make certain effectiveness evaluation.

Practical Approximation Approaches to Forecasting and Backtesting Initial Margin Requirements

As BCBS-IOSCO's introduction of mandatory bilateral initial margin is expected to transform the bilateral OTC market, there has been a great interest in the market to develop a model that can dynamically forecast initial margin for future horizons. In the context of counterparty credit risk management and regulatory capital calculation, the authors propose practical methods to simulate initial margin requirements, along with validation and historical backtesting approaches applicable to a broad class of initial margin simulation methods.

Multiperiod conditional valuation of barrier options with incomplete information

In this paper, we provide analytical valuation results for barrier options with three different types of information: continuous observations of the underlying asset value, delayed continuous observations and multiple discrete observations. Market, counterparty credit risk management and economic risk capital implications of the valuations with incomplete information are also discussed. We obtain precise analytical solutions using all the information generated by the previous discrete observations of the underlying asset, generated in a consistent fashion by path-dependent simulation. Our results have far-reaching implications for economic capital charges on path-dependent derivatives. We demonstrate that using non-conditional valuation mis-estimates the credit charges on the barrier options with respect to the precise credit charges computed with valuations conditional on the actual available information. The rest of this paper is organized as follows. Section 2 describes the classical case with full, continuous information about the underlying asset. Section 3 deals with the case of delayed continuous information. Section 4 investigates the case of discrete observations at multiple discrete times. Section 5 offers explicit closed-form solutions for the main types of single-barrier options with discrete information and parity results. Section 6 provides valuation results for double-barrier options. Section 7 presents risk management applications for foreign exchange (FX) barrier options, while section 8 concludes our paper.

Counterparty Credit Risk Management in Industrial Corporates

Ever since the financial crisis of the banking system of 2008 - 2010 the paradigm that deposits or other exposures towards major banks are safe has been fundamentally questioned. This put industrial corporates, who to support their business usually need to manage significant cash holdings or incur counterparty credit risk via derivatives, in the situation to develop or extend their resources for counterparty credit risk management. This paper provides a comprehensive overview over the practical issues into the subject benefitting largely from the findings of an interview series conducted with the respective heads of counterparty and customer credit risk management in the time period April - September 2011 of 25 large european enterprises with a large subset being members of the German DAX Index.

Optimal Funding Strategies for Counterparty Credit Risk Liabilities

The Dodd-Frank Act and the recently proposed Basel Committee regulatory framework for CCPs are a game changer for counterparty credit risk management. The practice of charging an upfront fee as a Credit Valuation Adjustment (CVA) to provision against counterparty credit risk liabilities is being abandoned as it was blamed for as much as two thirds of the losses recorded during the financial crisis. Instead, a key role will be played by margin financing, whereby periodically marked-to-market revolving lines of credit are used to cover margin variations on a cross-product basis. The emerging pay-as-you-go funding strategy for counterparty credit risk liabilities has a fair value equal to the CVA upfront fee but an entirely different risk profile. Using margin financing, the process for expected loss is locked at zero by construction and CVA volatility risk is passed on to the counterparties themselves. As a side effect, wealth is transferred from bankrupt entities to healthy ones. Moreover, in a Dodd-Frank world, there is no DVA because there is no counterparty credit risk and the paradoxes of DVA accounting and CSA discounting are removed. To further optimize the funding strategy, interest in flows from portfolios of margin revolvers can be redirected through securitization vehicles to a hierarchy of bond holders to which tranches of risk are apportioned. With this construction, banks can purchase nearly full counterparty default protection from capital markets at a fair cost equal to the CVA, the theoretical optimum. The only remaining risk is concentrated in the equity tranche as there could be a mismatch between interest inflows and outflows.

Hedge Funds, Financial Intermediation, and Systemic Risk

Hedge funds, with assets under management approaching an estimated $1.5 trillion in 2006, have become important players in the U.S. and global capital markets. These largely unregulated funds differ from other market participants in their use of a variety of complex trading strategies and instruments, in their liberal use of leverage, in their opacity to outsiders, and in their convex compensation structure. These differences can exacerbate potential market failures stemming from agency problems, externalities, and moral hazard. Counterparty credit risk management (CCRM) practices, used by financial institutions to assess credit risk and limit counterparty exposure, are the first line of defense against market disruptions with potential systemic consequences. This article examines how the unique nature of hedge funds may generate market failures that make counterparty credit risk for exposures to the funds intrinsically more difficult to manage, both for regulated institutions and for policymakers concerned with systemic risk. The authors acknowledge that various market failures, such as the events surrounding the 1998 collapse of hedge fund Long-Term Capital Management, may make CCRM imperfect. However, CCRM has improved significantly since then, and it remains the appropriate starting point for limiting the potential for hedge funds to generate systemic disruptions.

Toward a Better Estimation of Wrong‐Way Credit Exposure

In counterparty credit risk management for swaps, forwards, and other derivative contracts, it is recognized that most common applications of credit exposure measurement suffer from the bias that counterparty default is independent of the amount of exposure. Stress tests are often proposed to compensate for this bias, but these measures tend to be arbitrary and cannot be uniformly applied to setting prices and limits as readily as more standardized approaches. The author proposes a framework in which standard measures of counterparty exposure are conditioned on default probabilities. These conditional measures thus account for “rong way” exposures, but fit naturally into current applications.

Toward A Better Estimation of Wrong-Way Credit Exposure

In counterparty credit risk management for swaps, forwards, and other derivative contracts, it is recognized that most common applications of credit exposure measures suffer from the deficiency of assuming that counterparty default is independent of the amount exposed. Stress tests are often proposed to handle this deficiency, but stress test measures cannot be applied to pricing and limit setting in the same vein as the standard measures. We introduce a framework to condition standard measures of counterparty exposure on default. The conditional measures thus account for wrong-way exposures, but fit naturally into current applications.