ALM Model Validation
Inspire more confident risk assessment and strategic decision making at ALCO.
Confident decision-making at ALCO begins with an ALM model validation that unequivocally affirms that your model performs as intended and that your interest rate risk (IRR) assessment is accurate. ALM models are becoming increasingly more complex, being used for and more reliant on expanded data elements and dynamic assumption inputs, often driven by feeder models. DCG has the market insight, experience, and expertise to properly assess your model's vulnerabilities and deliver meaningful enhancements for all your ALM model uses.
“The best validations go beyond the technical review; they assess alternative methods and provide feedback for continued enhancements to your operation.”
Robust Data Testing
Clean, accurate, and appropriate data form the foundation of any risk model. DCG's robust ALM data validation process pinpoints specific data quality issues and uncovers underlying errors or shortcomings in the data feeding your ALM model that can materially affect the accuracy of your simulations.
Benchmarking and back-testing are some of the most effective ways to confirm that the aspects of your ALM model are performing as intended and producing reliable results. DCG's ALM model validation experts utilize a series of state-of-the-art proprietary and 3rd-party models and tools to benchmark key elements of your model, including your investment portfolio's cash flows and valuations under varying scenarios.
Keen Focus on Assumptions
The assumptions applied to ALM models have a significant impact on results and need to be meaningfully assessed as part an effective ALM model validation framework. In conjunction with the ALM model validation, DCG's experts will evaluate all of its key assumptions, how they are derived and applied to the model, along with how they are routinely supported, tested, and approved. This assessment can also be expanded to include limited- and full-scope validations of in-house and third-party assumption models.
Thorough Governance Assessments
Whether your ALM model is managed in-house or outsourced, it is essential to maintain robust processes and controls along with appropriately detailed model documentation. Additionally, the policies that govern your risk management practices should accurately reflect your operating philosophies and risk tolerances, ensure sufficient oversight and flexibility, and align with current regulatory expectations. DCG's ALM model validation includes a thorough governance review informed by decades of industry-leading IRR modeling, ALCO advisory, and examiner training.
Companion ALCO Performance Reviews
Even with good modeling, ALCOs can struggle to leverage their modeling results strategically. In some instances, results are not communicated well or not understood. In other cases, there are operational or cultural impediments that get in the way. Leveraging four decades of ALCO advisory experience, DCG's experts can independently assess your ALCO process and identify ways to optimize your team's strategic contribution. This assessment includes an in-depth review of your ALCO-related policies and procedures, reporting materials, approach to strategy formulation, decision making, and execution.
Optimized Model Use
ALM models have evolved into more than just a projection engine for earnings at risk and EVE analysis. Many institutions leverage the ALM model within their liquidity risk, credit risk, and stress testing programs. It is required that ALM models be validated for every use, so the potential exposure and need for specialized validation experts continues to grow. DCG has the experience to validate for each model use while providing perspectives that have implications for all uses. The potential risk exposure from these models is too great to take a narrow-sighted approach to your validation.
How We Help Clients
Examples of Client Results
Identified assumption deficiencies through model back-testing / performance monitoring, making IRR model results more accurate.
Identified IRR model being used inappropriately for liquidity forecasting, producing unreliable information to key decision-makers at $1.5B bank. Worked with bank to build new liquidity model, resulting in higher confidence in output.