ABSTRACT

Model risk as part of operational risk is a serious problem for financial institutions. As the pricing of derivatives as well as the computation of the market or credit risk of an institution depend on statistical models, the application of a wrong model can lead to a serious over- or underestimation of the institution's risk. Because the underlying data-generating process is unknown in practice, evaluating the model risk is a challenge. So far, definitions of model risk have been either application-oriented, including risk induced by the statistician rather than by the statistical model, or so research-oriented as to prove too abstract to be used in practice. They are especially prone not to be data driven. We introduce a data-driven notion of model risk that includes the features of the research-oriented approach, extending it by a statistical model-building procedure and thus compromising between the two definitions at hand. We further suggest the application of robust estimates to reduce the model risk, and advocate the application of stress tests with respect to the valuation of the portfolio.