Mudelirisk

Mudelirisk, also known as model risk, is the risk that the output of a model is inaccurate or misleading for decision making. It arises when a model’s structure, assumptions, data, or implementation do not adequately reflect reality, or when the model is used beyond its intended purpose.

Common sources of mudelirisk include poor data quality or insufficient data, incorrect model specification, errors in estimation or calibration, overfitting, model obsolescence, software bugs, and misapplication of model results. Other factors are changes in the environment that the model does not capture, or inadequate governance and change management.

In practice, mudelirisk appears in many domains, especially finance. Examples include pricing and valuation models, credit risk and underwriting models, market risk and economic capital models, forecasting and stress-testing tools, and risk-management dashboards. The risk can lead to mispriced products, underestimated capital needs, poor strategic decisions, or ineffective risk controls.

Management of mudelirisk relies on a formal model risk governance framework. Key elements are a model inventory with owner and purpose, independent model validation, and defined approval and change-control processes. Ongoing activities include backtesting and backtesting-based monitoring, out-of-sample testing, sensitivity and scenario analyses, benchmarking against alternatives, and documentation of assumptions and limitations. Regulators often require robust validation, governance, and disclosure to ensure that model risk is identified, quantified, and mitigated.

Regulatory guidance varies by jurisdiction but commonly emphasizes independent validation, risk appetite alignment, and adequate capital and controls to address model risk.