A sound and independent model validation is essential for running reliable and robust risk management systems and methodologies. A comprehensive model validation framework with strong governance ensures effective identification of potential model risks at Eurex Clearing AG.
Eurex Clearing commits itself to a regular and thorough validation of all risk models and model-related processes for its models. While model owners are ultimately responsible for model development and associated model risk, model validation acts as a second line of defence. To ensure independence from model development, model validation reports directly to the Chief Risk Officer of Eurex Clearing and has an alternative reporting line to the Chair of the Supervisory Board of Eurex Clearing.
Application of different validation instruments is governed by the model validation framework along with procedures for regular, ad-hoc validation and findings tracking. Throughout the year, the independent model validation uses a number of validation instruments to regularly validate the conceptual soundness of the frameworks and adequacy of the risk models. Once a year, an annual comprehensive validation is performed for all models in scope. In the annual validation report, all validation results obtained within the year are summarized and combined with a fundamental review of the methodology and the model parametrization. This yields an overall review of model performance and appropriateness. The report also assesses the effectiveness of processes and procedures relevant for managing model risks.
The adequacy of the initial margin is primarily validated by portfolio backtesting, which compares the initial margin with the actually realised profits and losses. The results are assessed using statistical tests. Validation at parameter level is performed by means of a parameter sensitivity analysis, which is conducted to determine the margin model’s response to changes in model parameters.
Results from backtesting and parameter sensitivity analysis are regularly reported to the Risk Committee in a form that does not breach confidentiality.
Haircut backtesting is in place to validate the adequacy of haircuts on cash and non-cash collateral on a monthly basis. Results are also assessed using statistical tests. Eligibility criteria for collateral acceptance are also reviewed.
With the stress testing validation, the appropriateness of the stress testing methodology with all its components such as scenarios, parameters and assumptions is performed on an annual basis. Specifically, the adequacy of the default fund calibration methodology is verified as a part of the stress testing validation. Anonymized results of the annual review are presented to the Risk Committee.
The performance of the liquidity forecasting model, which is used to predict Eurex Clearing’s liquidity needs and to evaluate available liquidity sources in case that one or more clearing members fail to fulfill their obligations, is assessed on an annual basis.
Credit, concentration and wrong-way risk
The conceptual soundness of the valuation models applied in Eurex Clearing’s CCP Risk Management Framework are regularly reviewed.
Model-related processes are operational business processes which outcome is of quantitative nature and used as a parameter in one of the risk models with a significant impact on the model output.
The data quality assurance process is crucial for Eurex Clearing’s model operation as it ensures high quality of input data. General and product-specific outlier detection and correction methods are applied to both raw market data and derived data.
Validation of the default management process aims to ensure that the conceptual design of the process is adequate.
Offset monitoring ensures that portfolio diversification benefits granted by the portfolio margining approach stay within regulatory required limits and that Eurex Clearing is not exposed to further potential risks by the margin reduction.